{"id":54563,"date":"2024-12-20T20:37:29","date_gmt":"2024-12-20T12:37:29","guid":{"rendered":"https:\/\/swarma.org\/?p=54563"},"modified":"2024-12-20T20:37:29","modified_gmt":"2024-12-20T12:37:29","slug":"%e5%85%b7%e8%ba%ab%e6%99%ba%e8%83%bd%e8%af%bb%e4%b9%a6%e4%bc%9a%e5%90%af%e5%8a%a8%ef%bc%9a%e8%b5%b0%e5%90%91%e7%8e%b0%e5%ae%9e%e4%b8%96%e7%95%8c%e7%9a%84%e4%b8%8b%e4%b8%80%e4%bb%a3ai%e7%b3%bb%e7%bb%9f","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=54563","title":{"rendered":"\u5177\u8eab\u667a\u80fd\u8bfb\u4e66\u4f1a\u542f\u52a8\uff1a\u8d70\u5411\u73b0\u5b9e\u4e16\u754c\u7684\u4e0b\u4e00\u4ee3AI\u7cfb\u7edf"},"content":{"rendered":"
\n
\n
\n
\n

<\/p>\n

Jeffrey Fisher for Quanta Magazine<\/span><\/p>\n<\/section>\n

\n
\n
\n
\n
\n
\n

<\/section>\n<\/section>\n
\n
\n

\u5bfc\u8bed<\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

\u96c6\u667a\u4ff1\u4e50\u90e8\u8054\u5408\u4e0a\u6d77\u4ea4\u901a\u5927\u5b66\u52a9\u7406\u6559\u6388\u674e\u6c38\u9732\u3001\u94f6\u6cb3\u901a\u7528\u673a\u5668\u4eba\u5408\u4f19\u4eba\u53f2\u96ea\u677e\u3001\u5357\u4eac\u5927\u5b66LAMDA\u7ec4\u535a\u58eb\u751f\u9648\u96c4\u8f89\u3001\u9999\u6e2f\u5927\u5b66\u5728\u8bfb\u535a\u58eb\u751f\u7a46\u5c27\uff0c\u5171\u540c\u53d1\u8d77\u9996\u5b63\u300c\u5177\u8eab\u667a\u80fd\u300d\u8bfb\u4e66\u4f1a\u3002\u8bfb\u4e66\u4f1a\u8ba1\u5212\u91c7\u7528\u201c\u81ea\u4e0b\u800c\u4e0a\u201d\u7684\u5c42\u7ea7\u7ed3\u6784\uff0c\u63a2\u8ba8\u56db\u4e2a\u6838\u5fc3\u6a21\u5757\uff1a\u786c\u4ef6\u7cfb\u7edf\uff08\u673a\u5668\u4eba\u672c\u4f53\u8bbe\u8ba1\uff09\uff0c\u6570\u636e\u3001\u4eff\u771f\u73af\u5883\u4e0eBenchmark\uff0c\u673a\u5668\u4eba\u5b66\u4e60\uff0c\u5177\u4f53\u573a\u666f\u4efb\u52a1\u3002\u5e0c\u671b\u901a\u8fc7\u91cd\u70b9\u8ba8\u8bba\u7ecf\u5178\u3001\u524d\u6cbf\u7684\u91cd\u8981\u6587\u732e\uff0c\u5e2e\u52a9\u5927\u5bb6\u66f4\u597d\u5730\u5b66\u4e60\u673a\u5668\u4eba\u4e0e\u5177\u8eab\u667a\u80fd\u6280\u672f\u524d\u6cbf\u6280\u672f\uff0c\u4e3a\u76f8\u5173\u9886\u57df\u7684\u7814\u7a76\u548c\u5e94\u7528\u63d0\u4f9b\u6d1e\u89c1\u3002<\/span><\/strong><\/p>\n


<\/span><\/strong><\/p>\n

\u8bfb\u4e66\u4f1a\u4ece2025\u5e741\u670819\u65e5\u5f00\u59cb\uff0c\u6bcf\u5468\u65e514:00-16:00\uff0c\u6301\u7eed\u65f6\u95f4\u9884\u8ba1 6-8 \u5468\u5de6\u53f3\u3002\u6bcf\u5468\u8fdb\u884c\u7ebf\u4e0a\u4f1a\u8bae\uff0c\u4e0e\u4e3b\u8bb2\u4eba\u7b49\u793e\u533a\u6210\u5458\u5f53\u9762\u4ea4\u6d41\uff0c\u4e4b\u540e\u53ef\u4ee5\u83b7\u5f97\u89c6\u9891\u56de\u653e\u6301\u7eed\u5b66\u4e60\u3002<\/span><\/strong><\/p>\n<\/section>\n

\n

<\/span><\/strong><\/p>\n<\/section>\n<\/section>\n

\n
<\/strong><\/strong><\/section>\n
<\/strong><\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
<\/section>\n<\/section>\n


<\/p>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

\u80cc\u666f\u4ecb\u7ecd<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n

\u4eba\u5de5\u667a\u80fd\u7684\u53d1\u5c55\u6b63\u9762\u4e34\u7740\u4e00\u4e2a\u6839\u672c\u6027\u7684\u6311\u6218\uff1a\u5982\u4f55\u8ba9AI\u7cfb\u7edf\u771f\u6b63\u7406\u89e3\u548c\u9002\u5e94\u7269\u7406\u4e16\u754c\u7684\u590d\u6742\u6027\u3002\u5c3d\u7ba1\u5728\u6570\u5b57\u9886\u57df\u53d6\u5f97\u4e86\u5de8\u5927\u7a81\u7834\uff0c\u4f46\u5728\u4e0e\u73b0\u5b9e\u4e16\u754c\u7684\u4ea4\u4e92\u65b9\u9762\uff0cAI\u7cfb\u7edf\u4ecd\u7136\u663e\u5f97\u5c40\u9650\u3002\u8fd9\u79cd\u5dee\u8ddd\u63ed\u793a\u4e86\u667a\u80fd\u7cfb\u7edf\u5728\u201c\u5177\u8eab\u6027\u201d\u65b9\u9762\u7684\u91cd\u8981\u7f3a\u5931\u2014\u2014\u5b83\u4eec\u7f3a\u4e4f\u901a\u8fc7\u7269\u7406\u4ea4\u4e92\u6765\u83b7\u53d6\u548c\u9a8c\u8bc1\u77e5\u8bc6\u7684\u80fd\u529b\u3002<\/span><\/p>\n


<\/span><\/p>\n

\u5177\u8eab\u667a\u80fd\uff08Embodied AI\uff09\u6b63\u662f\u4e3a\u4e86\u89e3\u51b3\u8fd9\u4e00\u6838\u5fc3\u95ee\u9898\u800c\u63d0\u51fa\u7684\u7814\u7a76\u65b9\u5411\u3002\u5b83\u8bd5\u56fe\u901a\u8fc7\u5c06AI\u7cfb\u7edf\u5d4c\u5165\u7269\u7406\u8f7d\u4f53\u4e2d\uff08\u5982\u673a\u5668\u4eba\uff09\uff0c\u4f7f\u5176\u80fd\u591f\u76f4\u63a5\u548c\u771f\u5b9e\u4e16\u754c\u611f\u77e5\u548c\u4ea4\u4e92\u3002\u8fd9\u4e00\u8fc7\u7a0b\u4e2d\uff0c\u6211\u4eec\u9762\u4e34\u7740\u51e0\u4e2a\u672c\u8d28\u6027\u7684\u6311\u6218\uff1a<\/span><\/section>\n
    \n
  1. \n

    \u590d\u6742\u6027\u6311\u6218\uff1a\u5982\u4f55\u5728\u8fd9\u79cd\u590d\u6742\u73af\u5883\u4e2d\u5b9e\u73b0\u7a33\u5b9a\u53ef\u9760\u7684\u611f\u77e5\u3001\u63a7\u5236\u548c\u51b3\u7b56\uff1f<\/span><\/p>\n<\/li>\n

  2. \n

    \u5b66\u4e60\u4e0e\u6cdb\u5316\uff1a\u5982\u4f55\u4ece\u6709\u9650\u7684\u7269\u7406\u4ea4\u4e92\u4e2d\u5b66\u4e60\u5e76\u6cdb\u5316\u5230\u65b0\u7684\u573a\u666f\uff1f<\/span><\/p>\n<\/li>\n

  3. \n

    \u591a\u5c3a\u5ea6\u6574\u5408\uff1a\u4ece\u5e95\u5c42\u7684\u4f20\u611f\u5668\u6570\u636e\u5230\u9ad8\u5c42\u7684\u4efb\u52a1\u89c4\u5212\uff0c\u5982\u4f55\u5b9e\u73b0\u8de8\u8d8a\u591a\u4e2a\u5c3a\u5ea6\u7684\u6709\u6548\u6574\u5408\uff1f<\/span><\/p>\n<\/li>\n

  4. \n

    \u7cfb\u7edf\u9c81\u68d2\u6027\uff1a\u5728\u5f00\u653e\u3001\u52a8\u6001\u3001\u4e0d\u786e\u5b9a\u7684\u73af\u5883\u4e2d\uff0c\u5982\u4f55\u4fdd\u8bc1\u7cfb\u7edf\u7684\u53ef\u9760\u6027\u548c\u5b89\u5168\u6027\uff1f<\/span><\/p>\n<\/li>\n

  5. \n

    \u667a\u80fd\u6d8c\u73b0\uff1a\u5177\u8eab\u7ecf\u9a8c\u5982\u4f55\u4fc3\u8fdb\u66f4\u9ad8\u5c42\u6b21\u667a\u80fd\u7684\u4ea7\u751f\uff1f\u7269\u7406\u4ea4\u4e92\u662f\u5426\u662f\u5b9e\u73b0\u771f\u6b63\u667a\u80fd\u7cfb\u7edf\u7684\u5fc5\u8981\u6761\u4ef6\uff1f<\/span><\/p>\n<\/li>\n<\/ol>\n


    <\/span><\/p>\n

    \u8fd9\u4e9b\u95ee\u9898\u7684\u63a2\u7d22\u9700\u8981\u8de8\u8d8a\u591a\u4e2a\u5b66\u79d1\u9886\u57df\uff0c\u4ece\u673a\u5668\u4eba\u5b66\u3001\u63a7\u5236\u8bba\u3001\u8ba4\u77e5\u79d1\u5b66\u5230\u673a\u5668\u5b66\u4e60\u3002\u7279\u522b\u662f\u5728\u5f53\u524d\u5927\u8bed\u8a00\u6a21\u578b\u84ec\u52c3\u53d1\u5c55\u7684\u80cc\u666f\u4e0b\uff0c\u5982\u4f55\u5c06\u8bed\u8a00\u7406\u89e3\u4e0e\u7269\u7406\u4ea4\u4e92\u80fd\u529b\u76f8\u7ed3\u5408\uff0c\u6210\u4e3a\u4e00\u4e2a\u6781\u5177\u524d\u666f\u7684\u65b0\u65b9\u5411\u3002<\/span><\/p>\n


    <\/span><\/p>\n


    <\/h2>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    \u8bfb\u4e66\u4f1a\u6846\u67b6<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/p>\n

    \u9488\u5bf9\u5177\u8eab\u667a\u80fd\u9996\u5b63\u8bfb\u4e66\u4f1a\uff0c\u6211\u4eec\u8bbe\u8ba1\u4e86\u4e00\u4e2a\u7cfb\u7edf\u5316\u7684\u5b66\u4e60\u6846\u67b6\uff0c\u5e0c\u671b\u5927\u5bb6\u5bf9\u5177\u8eab\u667a\u80fd\u8fd9\u4e2a\u9886\u57df\u6709\u5168\u5c40\u7684\u4e86\u89e3\u3002\u8fd9\u4e2a\u6846\u67b6\u91c7\u7528\u201c\u81ea\u4e0b\u800c\u4e0a\u201d\u7684\u5c42\u7ea7\u7ed3\u6784\uff0c\u6309\u7167\u6280\u672f\u6808\u4ece\u5e95\u5c42\u5230\u9ad8\u5c42\u4f9d\u6b21\u5305\u542b\uff1a\u786c\u4ef6\u7cfb\u7edf<\/span>\uff08\u673a\u5668\u4eba\u672c\u4f53\u8bbe\u8ba1\uff09<\/span>\u3001\u6570\u636e\u3001\u4eff\u771f\u73af\u5883\u4e0eBenchmark\u3001\u673a\u5668\u4eba\u5b66\u4e60\u3001\u4ee5\u53ca\u5177\u4f53\u573a\u666f\u4efb\u52a1\u3002\u8fd9\u79cd\u5212\u5206\u65e2\u53cd\u6620\u4e86\u6280\u672f\u4e4b\u95f4\u7684\u4f9d\u8d56\u5173\u7cfb\uff0c\u4e5f\u4f53\u73b0\u4e86\u4ece\u7269\u7406\u5b9e\u73b0\u5230\u667a\u80fd\u5e94\u7528\u7684\u81ea\u7136\u6f14\u8fdb\u8fc7\u7a0b\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u5176\u4e2d\uff0c\u5728\u6846\u67b6\u7684\u6838\u5fc3\u90e8\u5206\u2014\u673a\u5668\u4eba\u5b66\u4e60<\/span>\uff08robotics learning\uff09<\/span>\u4e2d\uff0c\u6211\u4eec\u4ece\u4e24\u4e2a\u4e92\u8865\u7684\u7ef4\u5ea6\u8fdb\u884c\u6df1\u5165\u63a2\u8ba8\uff1a<\/span><\/p>\n


    <\/span><\/p>\n

    \u7b2c\u4e00\u4e2a\u7ef4\u5ea6\u662f\u51b3\u7b56\u6846\u67b6\uff0c\u6309\u7167\u4fe1\u606f\u5904\u7406\u6d41\u7a0b\u5206\u4e3a\u611f\u77e5\u3001\u63a7\u5236\u548c\u89c4\u5212\u4e09\u4e2a\u5c42\u6b21\uff0c\u540c\u65f6\u4e5f\u8003\u8651\u7aef\u5230\u7aef\u7684\u65b9\u6cd5\u3002\u8fd9\u4e2a\u5212\u5206\u9075\u5faa\u201c\u4ece\u7269\u7406\u5230\u4efb\u52a1\u201d\u7684\u539f\u5219\u2014\u8d8a\u9760\u8fd1\u5de6\u4fa7\u8d8a\u8d34\u8fd1\u7269\u7406\u4e16\u754c\u7684\u539f\u59cb\u4fe1\u606f\uff0c\u8d8a\u9760\u53f3\u4fa7\u8d8a\u63a5\u8fd1\u6700\u7ec8\u8981\u5b8c\u6210\u7684\u76ee\u6807\u4efb\u52a1\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u7b2c\u4e8c\u4e2a\u7ef4\u5ea6\u662f\u5b66\u4e60\u65b9\u6cd5\u8bba\uff0c\u5305\u62ec\u4ece\u771f\u5b9e\u6570\u636e\u76f4\u63a5\u5b66\u4e60\u3001\u4ece\u4eff\u771f\u73af\u5883\u5b66\u4e60\u5e76\u8fc1\u79fb\u3001\u57fa\u4e8e\u4e16\u754c\u6a21\u578b\u5b66\u4e60\u3001\u5229\u7528\u591a\u6a21\u6001\u6570\u636e\u5b66\u4e60\u3001\u4ee5\u53ca\u4ece\u5176\u4ed6\u9886\u57df\u77e5\u8bc6\u8fc1\u79fb\u5b66\u4e60\u3002\u8fd9\u4e2a\u7ef4\u5ea6\u4f53\u73b0\u4e86\u4e00\u4e2a\u91cd\u8981\u7684\u6743\u8861\u2014\u4ece\u5de6\u5230\u53f3\uff0c\u6570\u636e\u7684\u83b7\u53d6\u96be\u5ea6\u9010\u6e10\u964d\u4f4e\uff0c\u4f46\u4e0e\u76ee\u6807\u573a\u666f\u7684\u5dee\u5f02\u4e5f\u9010\u6e10\u589e\u5927\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/section>\n


    <\/span><\/p>\n


    <\/span><\/section>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    \u53d1\u8d77\u4eba\u4ecb\u7ecd<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/p>\n

    <\/p>\n

    \u53f2\u96ea\u677e\u8001\u5e08\u662f\u94f6\u6cb3\u901a\u7528\u673a\u5668\u4eba\u5408\u4f19\u4eba\u3001\u7b97\u6cd5\u5de5\u7a0b\u8d1f\u8d23\u4eba\uff0cCCF\u667a\u80fd\u673a\u5668\u4eba\u4e13\u59d4\u4f1a\u5e38\u52a1\u59d4\u5458\uff0c\u590d\u65e6\u5927\u5b66\u535a\u58eb\u3002\u66fe\u4efb\u9ad8\u4ed9\u673a\u5668\u4eba\u7b97\u6cd5\u603b\u76d1\uff0c\u5386\u7ecf\u9ad8\u4ed9\u5546\u7528\u6e05\u6d01\u673a\u5668\u4eba\u5168\u7403\u5927\u89c4\u6a21\u90e8\u7f72\u8fc7\u7a0b\uff0c\u5e26\u961f\u653b\u514b\u8bf8\u591a\u7b97\u6cd5\u95ee\u9898\u3002\u66fe\u4efb\u82f1\u7279\u5c14\u7814\u7a76\u9662\u4e3b\u4efb\u7814\u7a76\u5458\uff0c\u4e3a\u82f1\u7279\u5c14\u81ea\u4e3b\u79fb\u52a8\u673a\u5668\u4eba\u8f6f\u4ef6\u5305\u7684\u4e3b\u8981\u7b97\u6cd5\u8d21\u732e\u8005\uff0cOpenLORIS\u7cfb\u5217\u6570\u636e\u96c6\u4f5c\u8005\uff0cIROS 2019 Lifelong Robotic Vision Challenge\u7ade\u8d5b\u4e3b\u5e2d\u3002\u53f2\u96ea\u677e\u535a\u58eb\u66fe\u4e8e\u4fe1\u53f7\u5904\u7406\u3001\u673a\u5668\u4eba\u5b66\u548c\u8ba1\u7b97\u673a\u89c6\u89c9\u7684\u9876\u7ea7\u5b66\u672f\u4f1a\u8bae\u4e0e\u671f\u520a\u53d1\u8868\u5341\u591a\u7bc7\u8bba\u6587\uff0c\u62e5\u6709\u5341\u591a\u9879\u673a\u5668\u4eba\u548c\u4eba\u5de5\u667a\u80fd\u65b9\u5411\u7684\u56fd\u9645\u4e13\u5229\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n

    \u9648\u96c4\u8f89\uff1a\u5357\u4eac\u5927\u5b66LAMDA\u7ec4\u535a\u58eb\u751f\uff0c\u5bfc\u5e08\u662f\u4fde\u626c\u6559\u6388\u3002\u7814\u7a76\u91cd\u70b9\u5728\u4e8e\u89e3\u51b3\u5f3a\u5316\u5b66\u4e60\u5728\u5728\u7ebf\u4ea4\u4e92\u6210\u672c\u654f\u611f\u7684\u771f\u5b9e\u5e94\u7528\u573a\u666f\u4e2d\u7684\u6311\u6218\u548c\u57fa\u4e8e\u5927\u6a21\u578b\u7684\u51b3\u7b56\u7814\u7a76\uff0c\u6d89\u53ca\u6280\u672f\u5305\u62ec\u79bb\u7ebf\u5f3a\u5316\u5b66\u4e60\uff0c\u4e16\u754c\u6a21\u578b\u5b66\u4e60\uff0csim2real \u8fc1\u79fb\uff0c\u56e0\u679c\u63a8\u65ad\uff0c\u57fa\u4e8e\u5927\u8bed\u8a00\u6a21\u578b\u7684\u51b3\u7b56\u548c\u51b3\u7b56\u5927\u6a21\u578b\u7b49\u3002\u76ee\u524d\u670910+\u7bc7\u8bba\u6587\u53d1\u8868\u5728NeurIPS\uff0cICML\uff0cICLR\uff0cTPAMI\u7b49\u9876\u4f1a\u4e0a\u3002\u5176\u7814\u7a76\u6210\u679c\u4e5f\u5728\u4e92\u8054\u7f51\uff0c\u5316\u5de5\u548c\u519b\u5de5\u7b49\u591a\u884c\u4e1a\u6210\u529f\u843d\u5730\u3002<\/span><\/p>\n

    https:\/\/xionghuichen.github.io\/<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n

    \u7a46\u5c27\uff1a\u9999\u6e2f\u5927\u5b66\u5728\u8bfb\u535a\u58eb\u751f\uff0c\u5e08\u4ece\u7f57\u5e73\u6559\u6388\u3002\u7814\u7a76\u65b9\u5411\uff1a\u5177\u8eab\u667a\u80fd\u3001\u5f3a\u5316\u5b66\u4e60\u3001\u673a\u5668\u4eba\u63a7\u5236\u548c\u81ea\u52a8\u9a7e\u9a76\u3002\u4e2a\u4eba\u4e3b\u9875\uff1ayaomarkmu.github.io\u3002\u5148\u540e\u5728NeurIPS, ICML, ICLR, CVPR\u7b49\u9876\u4f1a\u9876\u520a\u53d1\u8868\u8bba\u658720\u4f59\u7bc7\uff0c\u66fe\u83b7ICCAS2020\u5927\u4f1a\u6700\u4f18\u5b66\u751f\u8bba\u6587\u5956\uff0cIEEE  IV2021\u6700\u4f18\u5b66\u751f\u8bba\u6587\u63d0\u540d\u5956\u7b49\u591a\u9879\u5b66\u672f\u5956\u52b1\uff0c\u4e8e2021\u5e74\u5728\u6e05\u534e\u5927\u5b66\u53d6\u5f97\u7855\u58eb\u5b66\u4f4d\uff0c\u8363\u83b7\u9999\u6e2f\u535a\u58eb\u653f\u5e9c\u5956\u5b66\u91d1\uff0c\u9999\u6e2f\u5927\u5b66\u6821\u957f\u5956\u5b66\u91d1\uff0c\u56fd\u5bb6\u5956\u5b66\u91d1\uff0c\u6e05\u534e\u5927\u5b66\u4f18\u79c0\u7855\u58eb\u6bd5\u4e1a\u751f\uff0c\u6e05\u534e\u5927\u5b66\u4f18\u79c0\u7855\u58eb\u8bba\u6587\u5956\u7b49\u8363\u8a89\u79f0\u53f7\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/section>\n

    \u674e\u6c38\u9732\u8001\u5e08\u662f\u4e0a\u6d77\u4ea4\u5927\u52a9\u7406\u6559\u6388\uff0c\u535a\u5bfc\uff0c\u7814\u7a76\u5177\u8eab\u667a\u80fd\u3001\u89c6\u89c9\u63a8\u7406\u3001\u884c\u4e3a\u7406\u89e3\uff0c\u4ee3\u8868\u5de5\u4f5cHAKE\uff08\u5f15\u75281200+\uff0cGithub Star 1.5K+\uff0c\u5b98\u7f51\u5168\u7403\u8bbf\u95ee15\u4e07+\u6b21\uff09\u3001AlphaPose\uff08\u5f15\u7528500+\uff0cGithub Star 8K+\uff09\uff0c\u5927\u5e45\u63d0\u9ad8\u89c6\u89c9\u884c\u4e3a\u7406\u89e3\u7cbe\u5ea6\u3002\u53d1\u8868\u7814\u7a76\u6210\u679c40+\uff08TPAMI\u3001NeurIPS\u3001ICML\u3001CVPR\u3001ICCV\u3001ECCV\u3001IJCV\uff09\uff0c\u8c37\u6b4c\u5f15\u75282200+\uff0c\u4ed6\u5f15\u8d85100\u8bba\u65877\u7bc7\uff08\u4e00\u7bc7ESI\u6392\u540d\u524d\u5343\u5206\u4e4b\u516b\uff09\uff1b\u5f00\u6e90\u9879\u76ee20\u4f59\u9879\uff0c\u83b7Github star 1.2\u4e07+\u3002\u4efb\u4e0a\u6d77\u4ea4\u5927ACM\u73ed\u300a\u8ba1\u7b97\u673a\u89c6\u89c9\u300b\u3001AI\u73ed\u300a\u865a\u62df\u73b0\u5b9e\u300b\u8bfe\u7a0b\u6559\u5e08\uff0cNeurIPS 2024 Area Chair\uff0c\u591a\u79cdAI\u9876\u4f1a\u9876\u520a\u5ba1\u7a3f\u4eba\uff0cVALSE EAC\uff0c\u4e2d\u56fd\u4eba\u5de5\u667a\u80fd\u5b66\u4f1a-\u5177\u8eab\u667a\u80fd\u4e13\u59d4\u4f1a\u6267\u59d4\u3001\u79d8\u4e66\u5904\u6210\u5458\u3002<\/span><\/p>\n

    https:\/\/dirtyharrylyl.github.io\/<\/span><\/p>\n


    <\/p>\n


    <\/span><\/section>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    \u62a5\u540d\u53c2\u4e0e\u8bfb\u4e66\u4f1a<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/p>\n

    \u8fd0\u884c\u6a21\u5f0f<\/strong><\/span><\/h3>\n

    \u4ece2025\u5e741\u6708<\/strong>19<\/strong>\u65e5\u5f00\u59cb<\/strong>\uff0c\u6bcf\u5468\u65e5\u4e0b\u534814:00-16:00\uff0c\u6301\u7eed\u65f6\u95f4\u9884\u8ba1 6-8 \u5468\u5de6\u53f3\uff0c\u6309\u8bfb\u4e66\u4f1a\u6846\u67b6\u8bbe\u8ba1\uff0c\u6bcf\u5468\u8fdb\u884c\u7ebf\u4e0a\u4f1a\u8bae\uff0c\u4e0e\u4e3b\u8bb2\u4eba\u7b49\u793e\u533a\u6210\u5458\u5f53\u9762\u4ea4\u6d41\uff0c\u4f1a\u540e\u53ef\u4ee5\u83b7\u5f97\u89c6\u9891\u56de\u653e\u6301\u7eed\u5b66\u4e60\u3002<\/span><\/p>\n


    <\/h4>\n

    \u62a5\u540d\u65b9\u5f0f<\/strong><\/span><\/h4>\n
    \u7b2c\u4e00\u6b65\uff1a\u626b\u7801\u586b\u5199\u62a5\u540d\u4fe1\u606f\u3002<\/span><\/section>\n

    <\/p>\n

    https:\/\/pattern.swarma.org\/study_group\/58?from=wechat<\/span><\/section>\n
    \u626b\u7801\u62a5\u540d\uff08\u53ef\u5f00\u53d1\u7968\uff09<\/span><\/section>\n


    <\/span><\/p>\n

    \u7b2c\u4e8c\u6b65\uff1a\u586b\u5199\u4fe1\u606f\u540e\uff0c\u4ed8\u8d39\u62a5\u540d\u3002<\/span><\/section>\n
    \u5982\u9700\u7528\u652f\u4ed8\u5b9d\u652f\u4ed8\uff0c\u8bf7\u5728PC\u7aef\u8fdb\u5165\u8bfb\u4e66\u4f1a\u9875\u9762\u62a5\u540d\u652f\u4ed8\uff1a<\/em><\/span><\/section>\n
    \u7b2c\u4e09\u6b65\uff1a\u6dfb\u52a0\u8fd0\u8425\u8d1f\u8d23\u4eba\u5fae\u4fe1\uff0c\u83b7\u53d6\u6240\u6709\u63a8\u8350\u8bba\u6587\u8d44\u6e90\u5305\uff0c\u62c9\u5165\u5bf9\u5e94\u4e3b\u9898\u7684\u8bfb\u4e66\u4f1a\u793e\u533a\uff08\u5fae\u4fe1\u7fa4\uff09\u3002<\/span><\/section>\n

    PS\uff1a\u4e3a\u786e\u4fdd\u4e13\u4e1a\u6027\u548c\u8ba8\u8bba\u7684\u805a\u7126\uff0c\u672c\u8bfb\u4e66\u4f1a\u8c22\u7edd\u8131\u79bb\u8bfb\u4e66\u4f1a\u4e3b\u9898\u548c\u590d\u6742\u79d1\u5b66\u95ee\u9898\u672c\u8eab\u7684\u7a7a\u6cdb\u7684\u54f2\u5b66\u548c\u601d\u8fa8\u5f0f\u8ba8\u8bba\uff1b\u5982\u679c\u51fa\u73b0\u8ba8\u8bba\u5185\u5bb9\u4e0d\u7b26\u5408\u8981\u6c42\u3001\u7ecf\u63d0\u9192\u65e0\u6548\u8005\uff0c\u4f1a\u88ab\u79fb\u9664\u7fa4\u804a\u5e76\u5bf9\u672a\u53c2\u4e0e\u90e8\u5206\u9000\u8d39\u3002<\/strong><\/span><\/p>\n


    <\/span><\/h3>\n

    \u52a0\u5165\u793e\u533a\u540e\u53ef\u4ee5\u83b7\u5f97\u7684\u8d44\u6e90\uff1a<\/span><\/strong><\/span><\/h3>\n
    \u5b8c\u6574\u6743\u9650\uff0c\u5305\u62ec\u7ebf\u4e0a\u95ee\u7b54\u3001\u5f55\u64ad\u56de\u770b\u3001\u8d44\u6599\u5171\u4eab\u3001\u793e\u7fa4\u4ea4\u6d41\u3001\u4fe1\u606f\u540c\u6b65\u3001\u5171\u521b\u4efb\u52a1\u83b7\u53d6\u79ef\u5206\u7b49<\/strong>\u3002<\/span><\/section>\n

    <\/p>\n


    <\/span><\/h3>\n

    \u53c2\u4e0e\u5171\u521b\u4efb\u52a1\u83b7\u53d6\u79ef\u5206\uff0c\u5171\u5efa\u5b66\u672f\u793e\u533a\uff1a<\/span><\/strong><\/span><\/h3>\n

    \u8bfb\u4e66\u4f1a\u91c7\u7528\u5171\u5b66\u5171\u7814\u673a\u5236\uff0c\u6210\u5458\u901a\u8fc7\u5185\u5bb9\u5171\u521b\u83b7\u79ef\u5206\uff08\u5b57\u5e55\u4fee\u6539\u3001\u8bfb\u4e66\u4f1a\u7b14\u8bb0\u3001\u8bba\u6587\u901f\u9012\u3001\u516c\u4f17\u53f7\u6587\u7ae0\u3001\u96c6\u667a\u767e\u79d1\u3001\u8bba\u6587\u89e3\u8bfb\u7b49\u5171\u521b\u4efb\u52a1<\/strong>\uff09\uff0c\u79ef\u5206\u7b26\u5408\u6761\u4ef6\u5373\u53ef\u9000\u8d39\u3002\u53d1\u8d77\u4eba\u548c\u4e3b\u8bb2\u4eba\u540c\u6837\u9075\u5faa\u6b64\u673a\u5236\uff0c\u65e0\u989d\u5916\u91d1\u94b1\u6fc0\u52b1\u3002<\/span><\/p>\n

    <\/section>\n

    PS\uff1a\u5177\u4f53\u53c2\u4e0e\u65b9\u5f0f\u53ef\u4ee5\u52a0\u5165\u8bfb\u4e66\u4f1a\u540e\u67e5\u770b\u5bf9\u5e94\u7684\u5171\u521b\u4efb\u52a1\u5217\u8868\uff0c\u9886\u53d6\u4efb\u52a1\uff0c\u4e0e\u8fd0\u8425\u8d1f\u8d23\u4eba\u6c9f\u901a\u8be6\u60c5\uff0c\u4e0a\u8ff0\u89c4\u5219\u7684\u6700\u7ec8\u89e3\u91ca\u6743\u5f52\u96c6\u667a\u4ff1\u4e50\u90e8\u6240\u6709\u3002<\/span><\/p>\n


    <\/span><\/h1>\n


    <\/span><\/p>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    Embodied AI topic \u5217\u8868<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n

    <\/span><\/section>\n
    \u9605\u8bfb\u6750\u6599\u8f83\u957f\uff0c\u4e3a\u4e86\u66f4\u597d\u7684\u9605\u8bfb\u4f53\u9a8c\uff0c\u5efa\u8bae\u60a8\u524d\u5f80\u96c6\u667a\u6591\u56fe\u6c89\u6d78\u5f0f\u9605\u8bfb\uff0c\u5e76\u53ef\u6536\u85cf\u611f\u5174\u8da3\u7684\u8bba\u6587\u3002<\/span><\/section>\n
    <\/section>\n
    \u8bfb\u4e66\u4f1a\u6587\u732e\u9605\u8bfb\u6e05\u5355<\/span><\/section>\n

    <\/span><\/section>\n
    \u8bf4\u660e\uff1a\u8fd9\u6b21\u4e3b\u8981\u662f\u7ed9\u5927\u5bb6\u63a8\u8350\u4ee5\u76f8\u5173\u7efc\u8ff0\u53c2\u8003\u6587\u732e\u4e3a\u4e3b\uff0c\u5177\u4f53\u6bcf\u6b21\u5206\u4eab\u7684\u6587\u732e\uff0c\u6211\u4eec\u4f1a\u5728\u540e\u9762\u5206\u4eab\u7684\u65f6\u5019\u9010\u6b65\u6574\u7406\u51fa\u6765\u3002<\/span><\/section>\n

    <\/span><\/section>\n
    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/section>\n

    \u63a8\u8350\u7406\u7531\uff1a\u8fd9\u662f\u975e\u5e38\u91cd\u8981\u7684\u7efc\u8ff0\u4e4b\u4e00\uff0c\u68b3\u7406\u7684\u6846\u67b6\u53c2\u8003\u8fd9\u7bc7\u6587\u732e\uff0c\u6240\u4ee5\u4e0b\u9762\u591a\u4e2a\u5730\u65b9\u90fd\u6709\u6d89\u53ca\uff0c\u8fd9\u662f\u8fd9\u7bc7\u7efc\u8ff0\u5bf9\u5e94\u6574\u7406\u7684github\u6587\u732e\uff1a<\/span><\/p>\n

    https:\/\/github.com\/HCPLab-SYSU\/Embodied_AI_Paper_List<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h2>\n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n<\/section>\n
    \n
    \u786c\u4ef6\u672c\u4f53<\/strong><\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

    <\/p>\n


    <\/span><\/p>\n

    \u673a\u5668\u4eba\u672c\u4f53\u4f5c\u4e3a\u5177\u8eab\u667a\u80fd\u7684\u7269\u7406\u8f7d\u4f53\uff0c\u5176\u5f62\u6001\u8bbe\u8ba1\u3001\u6750\u6599\u548c\u529f\u80fd\u5b9e\u73b0\u76f4\u63a5\u51b3\u5b9a\u4e86\u7cfb\u7edf\u7684\u64cd\u4f5c\u80fd\u529b\u3002\u5728\u5404\u7c7b\u673a\u5668\u4eba\u786c\u4ef6\u4e2d\uff0c\u6211\u4eec\u4f1a\u91cd\u70b9\u5206\u4eab\u7075\u5de7\u624b\uff08Dexterous Hand\uff09\u7684\u76f8\u5173\u7814\u7a76\uff0c\u4f5c\u4e3a\u673a\u5668\u4eba\u4e0e\u7269\u7406\u4e16\u754c\u8fdb\u884c\u590d\u6742\u4ea4\u4e92\u7684\u6838\u5fc3\u6267\u884c\u5668\uff0c\u7075\u5de7\u624b\u7684\u7814\u7a76\u9762\u4e34\u4e09\u4e2a\u6839\u672c\u6027\u7684\u79d1\u5b66\u95ee\u9898\uff1a\u5982\u4f55\u5b9e\u73b0\u7c7b\u4eba\u6c34\u5e73\u7684\u89e6\u89c9\u611f\u77e5\u80fd\u529b\u3001\u5982\u4f55\u89e3\u51b3\u9ad8\u7ef4\u5ea6\u52a8\u4f5c\u7a7a\u95f4\u4e0b\u7684\u89c4\u5212\u4e0e\u63a7\u5236\u3001\u4ee5\u53ca\u5982\u4f55\u6784\u5efa\u53ef\u6cdb\u5316\u7684\u64cd\u4f5c\u77e5\u8bc6\u8868\u5f81\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h2>\n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n<\/section>\n
    \n
    \u6570\u636e\u96c6\u3001benchmark\u4e0e\u4eff\u771f<\/span><\/strong><\/section>\n
    <\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

    <\/p>\n


    <\/span><\/p>\n

    \u5728\u5177\u8eab\u667a\u80fd\u9886\u57df\uff0c\u6570\u636e\u96c6\u3001Benchmark \u548c\u4eff\u771f\u4e09\u8005\u7d27\u5bc6\u5173\u8054\uff0c\u5171\u540c\u6784\u6210\u7814\u7a76\u7684\u57fa\u7840\u6846\u67b6\u3002\u6570\u636e\u96c6\u63d0\u4f9b\u8bad\u7ec3\u548c\u8bc4\u4f30\u7684\u591a\u6a21\u6001\u6570\u636e\uff0c\u4eff\u771f\u73af\u5883\u751f\u6210\u6570\u636e\u5e76\u6a21\u62df\u771f\u5b9e\u4e16\u754c\u573a\u666f\uff0c\u800c Benchmark \u5219\u901a\u8fc7\u6807\u51c6\u5316\u4efb\u52a1\u548c\u8bc4\u4ef7\u6307\u6807\u63a8\u52a8\u6280\u672f\u7684\u5bf9\u6bd4\u4e0e\u8fdb\u6b65\u3002\u5b83\u4eec\u89e3\u51b3\u4e86\u6570\u636e\u7a00\u7f3a\u3001\u7814\u7a76\u6807\u51c6\u5316\u548c\u5b9e\u9a8c\u53ef\u63a7\u6027\u7684\u95ee\u9898\u3002\u7136\u800c\uff0c\u8be5\u9886\u57df\u4ecd\u9762\u4e34\u6570\u636e\u96c6\u8d28\u91cf\u4e0e\u591a\u6837\u6027\u3001\u4eff\u771f\u4e0e\u73b0\u5b9e\u7684\u5dee\u8ddd\uff08Sim-to-Real Gap\uff09\u3001\u4efb\u52a1\u590d\u6742\u6027\u4e0e\u591a\u6837\u6027\uff0c\u4ee5\u53ca\u7edf\u4e00\u6027\u4e0e\u53ef\u6269\u5c55\u6027\u7b49\u5173\u952e\u6311\u6218\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Wang, Zipeng, et al. “All Robots in One: A New Standard and Unified Dataset for Versatile, General-purpose Embodied Agents.” arXiv preprint arXiv:2408.10899 (2024).<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u63d0\u51faARIO\u6807\u51c6\uff0c\u662f\u9996\u4e2a\u7edf\u4e00\u7684\u5177\u8eab\u667a\u80fd\u6570\u636e\u96c6\u6807\u51c6\uff0c\u5305\u542b\u5168\u9762\u7684\u4f20\u611f\u5668\u6a21\u6001\u548c\u673a\u5668\u4eba\u5e73\u53f0\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Kulkarni, Mihir, Theodor JL Forgaard, and Kostas Alexis. “Aerial Gym–Isaac Gym Simulator for Aerial Robots.” arXiv preprint arXiv:2305.16510<\/em> (2023).https:\/\/arxiv.org\/abs\/2305.16510<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u63d0\u51fa\u7684Aerial Gym\u662f\u4e00\u4e2a\u7a81\u7834\u6027\u7684\u7a7a\u4e2d\u673a\u5668\u4eba\u4eff\u771f\u5e73\u53f0\uff0c\u5176\u9ad8\u5ea6\u5e76\u884c\u5316\u7279\u6027\u4f7f\u5176\u80fd\u540c\u65f6\u6a21\u62df\u6570\u767e\u4e07\u67b6\u591a\u65cb\u7ffc\u98de\u884c\u5668\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Nvidia isaac sim: Robotics simulation and synthetic data<\/strong>, NVIDIA, 2023 https:\/\/developer.nvidia.com\/isaac\/sim<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aNVIDIA\u5f00\u53d1\u7684\u9ad8\u6027\u80fd\u4eff\u771f\u5e73\u53f0\uff0c\u63d0\u4f9b\u9ad8\u4fdd\u771f\u7269\u7406\u4eff\u771f\u3001\u5b9e\u65f6\u5149\u7ebf\u8ffd\u8e2a\u548c\u4e30\u5bcc\u7684\u673a\u5668\u4eba\u6a21\u578b\u5e93\uff0c\u652f\u6301\u6df1\u5ea6\u5b66\u4e60\u548c\u5927\u89c4\u6a21\u5e76\u884c\u8ba1\u7b97\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Genesis: A Generative and Universal Physics Engine for Robotics and Beyond\uff1bGitHub\uff1ahttps:\/\/github.com\/Genesis-Embodied-AI\/Genesis \u63a8\u8350\u7406\u7531\uff1a\u6700\u65b0\u53d1\u5e03\u7684Genesis\u751f\u6210\u5f0f\u7269\u7406\u5f15\u64ce\u4ee3\u8868\u4e86\u7269\u7406\u6a21\u62df\u9886\u57df\u7684\u91cd\u5927\u7a81\u7834\u3002Genesis\u662f\u4e00\u4e2a\u5168\u9762\u7684\u7269\u7406\u6a21\u62df\u5e73\u53f0\uff0c\u4e13\u4e3a\u901a\u7528\u673a\u5668\u4eba\u3001\u5177\u8eabAI\u548c\u7269\u7406AI\u5e94\u7528\u8bbe\u8ba1\uff0c\u8fd9\u662f\u4e00\u4e2a\u5386\u65f624\u4e2a\u6708\u3001\u753120\u591a\u4e2a\u5b66\u672f\u548c\u4ea7\u4e1a\u673a\u6784\u5171\u540c\u7814\u53d1\u7684\u9879\u76ee\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Mu, Yao, et al. “Robotwin: Dual-arm robot benchmark with generative digital twins (early version).” arXiv preprint arXiv:2409.02920 (2024).https:\/\/arxiv.org\/abs\/2409.02920<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u63d0\u51fa\u4e86\u4e00\u4e2a\u7ed3\u5408\u771f\u5b9e\u9065\u64cd\u4f5c\u6570\u636e\u548c\u6570\u5b57\u5b6a\u751f\u751f\u6210\u6570\u636e\u7684\u521b\u65b0\u57fa\u51c6\uff0c\u4e13\u6ce8\u4e8e\u53cc\u81c2\u673a\u5668\u4eba\u5de5\u5177\u4f7f\u7528\u548c\u4eba\u673a\u4ea4\u4e92\uff0c\u662f\u5177\u8eab\u667a\u80fd\u9886\u57df benchmark \u7814\u7a76\u7684\u91cd\u8981\u53c2\u8003\u6587\u732e\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n<\/section>\n
    \n
    \u673a\u5668\u4eba\u5b66\u4e60\uff08Robotics learning\uff09<\/span><\/strong><\/section>\n
    <\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

    <\/p>\n


    <\/span><\/p>\n

    \u673a\u5668\u4eba\u5b66\u4e60\u662f\u5177\u8eab\u667a\u80fd\u5b9e\u73b0\u4ece\u611f\u77e5\u5230\u6267\u884c\u5b8c\u6574\u95ed\u73af\u7684\u6838\u5fc3\u9014\u5f84\u3002\u5728\u5b66\u4e60\u67b6\u6784\u4e0a\uff0c\u901a\u8fc7\u591a\u6a21\u6001\u611f\u77e5\u3001\u89c4\u5212\u51b3\u7b56\u548c\u63a7\u5236\u6267\u884c\u7b49\u673a\u5236\uff0c\u8ba9\u667a\u80fd\u4f53\u5728\u7269\u7406\u4e16\u754c\u4e2d\u83b7\u5f97\u7c7b\u4eba\u7684\u4ea4\u4e92\u80fd\u529b\u3002\u5728\u5b66\u4e60\u8303\u5f0f\u4e0a\uff0c\u5305\u62ec\u4ece\u771f\u5b9e\u6570\u636e\u76f4\u63a5\u5b66\u4e60\u3001\u901a\u8fc7\u4eff\u771f\u5230\u73b0\u5b9e\u8fc1\u79fb\u5b66\u4e60\u3001\u57fa\u4e8e\u4e16\u754c\u6a21\u578b\u7684\u5b66\u4e60\u3001\u8de8\u5177\u8eab\u5b66\u4e60\u4ee5\u53ca\u4ece\u5176\u4ed6\u9886\u57df\u77e5\u8bc6\u8fc1\u79fb\u5b66\u4e60\u7b49\u591a\u79cd\u65b9\u5f0f\u3002\u968f\u7740Vision-Language-Action\u5927\u6a21\u578b\u7684\u53d1\u5c55\uff0c\u673a\u5668\u4eba\u5b66\u4e60\u6b63\u4ece\u4f20\u7edf\u7684\u4efb\u52a1\u7279\u5b9a\u8303\u5f0f\u5411\u901a\u7528\u5316\u3001\u7aef\u5230\u7aef\u7684\u65b9\u5411\u6f14\u8fdb\uff0c\u8fd9\u4e9b\u5b66\u4e60\u8303\u5f0f\u4e5f\u5728\u4e0e\u5927\u6a21\u578b\u6df1\u5ea6\u878d\u5408\uff0c\u4ee5\u63d0\u5347\u7cfb\u7edf\u7684\u6cdb\u5316\u80fd\u529b\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Ma, Yueen, et al. “A Survey on Vision-Language-Action Models for Embodied AI.” arXiv preprint arXiv:2405.14093<\/em> (2024).https:\/\/arxiv.org\/abs\/2405.14093<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u8fd9\u7bc7\u6587\u732e\u7cfb\u7edf\u6027\u5730\u7efc\u8ff0\u4e86\u89c6\u89c9-\u8bed\u8a00-\u52a8\u4f5c\u6a21\u578b\uff08VLAs\uff09\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u91cd\u70b9\u63a2\u8ba8\u5176\u5728\u5177\u8eab\u667a\u80fd\u4e2d\u4f5c\u4e3a\u591a\u6a21\u6001\u57fa\u7840\u6a21\u578b\u7684\u6838\u5fc3\u4f5c\u7528\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h3>\n

    \u5b66\u4e60\u67b6\u6784<\/span><\/strong><\/span><\/h3>\n


    <\/span><\/h4>\n

    \u611f\u77e5\uff08perception\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/p>\n

    \u611f\u77e5\u662f\u6307\u667a\u80fd\u4f53\u901a\u8fc7\u591a\u6a21\u6001\u4f20\u611f\u5668\u83b7\u53d6\u73af\u5883\u4fe1\u606f\u5e76\u8fdb\u884c\u7406\u89e3\u548c\u63a8\u7406\u7684\u80fd\u529b\uff0c\u6db5\u76d63D\u573a\u666f\u7406\u89e3\u3001\u52a8\u6001\u73af\u5883\u4e2d\u7684\u5b9e\u65f6\u611f\u77e5\u4e0e\u63a8\u7406\u3001\u591a\u6a21\u6001\u4fe1\u606f\u6574\u5408\u7b49\u65b9\u9762\u3002\u5bf9\u4e8e\u611f\u77e5\u8fd9\u4e2a\u8bdd\u9898\u6765\u8bf4\uff0c\u6700\u91cd\u8981\u7684\u95ee\u9898\u4e4b\u4e00\u662f\u591a\u4f20\u611f\u5668\u6570\u636e\u5982\u4f55\u878d\u5408\u5e76\u4ea7\u751f\u7a33\u5b9a\u53ef\u9760\u7684\u573a\u666f\u7406\u89e3\uff1f\u4e3b\u8981\u6311\u6218\u5305\u62ec\u52a8\u6001\u73af\u5883\u4e2d\u7684\u611f\u77e5\u9c81\u68d2\u6027\u3001\u591a\u6a21\u6001\u611f\u77e5\u7684\u6574\u5408\u3001\u4e3b\u52a8\u63a2\u7d22\u7b56\u7565\u7684\u8bbe\u8ba1\uff0c\u4ee5\u53ca\u611f\u77e5\u4e0e\u51b3\u7b56\u7684\u534f\u540c\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u7cfb\u7edf\u603b\u7ed3\u4e86\u5177\u8eab\u611f\u77e5\u7684\u5173\u952e\u6280\u672f\uff08\u5982\u89c6\u89c9SLAM\u30013D\u573a\u666f\u7406\u89e3\uff09\u548c\u672a\u6765\u65b9\u5411\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u63a7\u5236\u5668\uff08controller\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    \u63a7\u5236\u5668\u662f\u6307\u8d1f\u8d23\u5c06\u9ad8\u5c42\u4efb\u52a1\u89c4\u5212\u8f6c\u5316\u4e3a\u5177\u4f53\u673a\u5668\u4eba\u52a8\u4f5c\u7684\u4f4e\u5c42\u6b21\u63a7\u5236\u7cfb\u7edf\uff0c\u4e3b\u8981\u5305\u62ec\u5173\u8282\u4f4d\u7f6e\u63a7\u5236\u3001\u672b\u7aef\u6267\u884c\u5668\u59ff\u6001\u63a7\u5236\u7b49\u5177\u4f53\u52a8\u4f5c\u7684\u6267\u884c\u3002\u8fd9\u90e8\u5206\u6700\u91cd\u8981\u7684\u79d1\u5b66\u95ee\u9898\u662f\uff1a\u5982\u4f55\u5b9e\u73b0\u4ece\u9ad8\u5c42\u6307\u4ee4\u5230\u5e95\u5c42\u52a8\u4f5c\u7684\u53ef\u9760\u6620\u5c04\uff1f\u5176\u4e3b\u8981\u6311\u6218\u5305\u62ec\u5982\u4f55\u5728\u52a8\u6001\u73af\u5883\u4e2d\u5b9e\u73b0\u9ad8\u7cbe\u5ea6\u63a7\u5236\u3001\u63d0\u5347\u6570\u636e\u6548\u7387\u4e0e\u6cdb\u5316\u80fd\u529b\u3001\u878d\u5408\u591a\u6a21\u6001\u611f\u77e5\u4fe1\u606f\uff0c\u4ee5\u53ca\u89e3\u51b3\u4eff\u771f\u5230\u73b0\u5b9e\u7684\u8fc1\u79fb\u95ee\u9898\uff08Sim-to-Real Gap\uff09\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Xu, Zhiyuan, et al. “A survey on robotics with foundation models: toward embodied ai.” arXiv preprint arXiv:2402.02385<\/em> (2024).https:\/\/arxiv.org\/abs\/2402.02385<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u8be5\u7efc\u8ff0\u5168\u9762\u68b3\u7406\u4e86\u5177\u8eab\u667a\u80fd\u9886\u57df\u4e2d\u63a7\u5236\u7cfb\u7edf\u7684\u53d1\u5c55\u8109\u7edc\uff0c\u7279\u522b\u662f\u8be6\u7ec6\u5206\u6790\u4e86\u57fa\u4e8e\u5f3a\u5316\u5b66\u4e60\u3001\u6a21\u4eff\u5b66\u4e60\u7b49\u65b9\u6cd5\u7684\u63a7\u5236\u7b56\u7565\u5b66\u4e60\u6846\u67b6\uff08\u5982RT-1\u3001RT-2\u7b49\uff09\uff0c\u5e76\u7cfb\u7edf\u6027\u5730\u8ba8\u8bba\u4e86\u5f53\u524d\u9762\u4e34\u7684\u6311\u6218\u548c\u672a\u6765\u53d1\u5c55\u65b9\u5411\uff0c\u5bf9\u63a7\u5236\u5668\u7814\u7a76\u5177\u6709\u91cd\u8981\u7684\u6307\u5bfc\u610f\u4e49\u3002<\/span><\/p>\n


    <\/p>\n

    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u7efc\u8ff0\u6587\u7ae0\uff0c\u68b3\u7406\u4e86\u5177\u8eab\u667a\u80fd\uff08Embodied AI\uff09\u9886\u57df\u7684\u6838\u5fc3\u65b9\u5411\uff0c\u5305\u62ec\u611f\u77e5\u3001\u63a7\u5236\u3001\u51b3\u7b56\u4ee5\u53ca\u4eff\u771f\u7b49\u5173\u952e\u73af\u8282\uff0c\u5168\u9762\u603b\u7ed3\u4e86\u591a\u6a21\u6001\u5b66\u4e60\u3001\u4e3b\u52a8\u63a2\u7d22\u3001\u6570\u636e\u751f\u6210\u3001\u4eff\u771f\u4e0e\u73b0\u5b9e\u878d\u5408\u7b49\u65b9\u6cd5\uff0c\u5e76\u63a2\u8ba8\u4e86\u9886\u57df\u672a\u6765\u7684\u79d1\u5b66\u6311\u6218\u548c\u7814\u7a76\u8d8b\u52bf\u3002\u4e5f\u6709\u5bf9\u4e8e\u63a7\u5236\u5668\u7684\u5168\u9762\u7684\u7cfb\u7edf\u7684\u68b3\u7406\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u89c4\u5212\uff08planning\uff09<\/span><\/strong><\/span><\/h4>\n

    <\/span><\/section>\n

    \u5728\u5177\u8eab\u667a\u80fd\u9886\u57df\uff0c\u89c4\u5212\u5206\u4e3a\u9ad8\u5c42\u4efb\u52a1\u89c4\u5212\uff08Task Planning\uff09\u548c\u4f4e\u5c42\u52a8\u4f5c\u89c4\u5212\uff08Action Planning\uff09\u4e24\u4e2a\u5c42\u6b21\u3002Task Planning\u8d1f\u8d23\u5c06\u590d\u6742\u62bd\u8c61\u4efb\u52a1\u5206\u89e3\u4e3a\u5177\u4f53\u53ef\u6267\u884c\u7684\u5b50\u4efb\u52a1\u5e8f\u5217\uff0c\u800cAction Planning\u5219\u8d1f\u8d23\u5c06\u8fd9\u4e9b\u5b50\u4efb\u52a1\u8f6c\u5316\u4e3a\u5177\u4f53\u7684\u63a7\u5236\u52a8\u4f5c\u3002\u8fd9\u6d89\u53ca\u591a\u4e2a\u5173\u952e\u6311\u6218\uff1a\u5305\u62ec\u957f\u671f\u4efb\u52a1\u7684\u8fde\u7eed\u6267\u884c\u3001\u5bf9\u52a8\u6001\u73af\u5883\u53d8\u5316\u7684\u5b9e\u65f6\u9002\u5e94\u3001\u4efb\u52a1\u5206\u89e3\u4e0e\u5177\u4f53\u6267\u884c\u7684\u6709\u6548\u8854\u63a5\u3001\u591a\u6a21\u6001\u4fe1\u606f\uff08\u89c6\u89c9\u3001\u8bed\u8a00\u7b49\uff09\u7684\u6574\u5408\u5229\u7528\uff0c\u4ee5\u53ca\u89c4\u5212\u7cfb\u7edf\u5728\u65b0\u573a\u666f\u548c\u65b0\u4efb\u52a1\u4e2d\u7684\u6cdb\u5316\u80fd\u529b\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u4e00\u5f00\u59cb\u7684\u63a8\u8350\u7efc\u8ff0\u6587\u7ae0\uff0c\u5176\u4e2d\u6709\u7cfb\u7edf\u7684\u68b3\u7406planning\u7684\u6a21\u5757\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u7aef\u5230\u7aef\uff08end-to-end\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/p>\n

    \u7aef\u5230\u7aef\u7cfb\u7edf\u81f4\u529b\u4e8e\u5b9e\u73b0\u4ece\u611f\u77e5\u8f93\u5165\uff08\u89c6\u89c9\u3001\u8bed\u8a00\u6307\u4ee4\u7b49\uff09\u76f4\u63a5\u6620\u5c04\u5230\u673a\u5668\u4eba\u63a7\u5236\u52a8\u4f5c\u7684\u5b8c\u6574\u95ed\u73af\u80fd\u529b\uff0c\u8fd9\u4e00\u65b9\u5411\u968f\u7740Vision-Language-Action (VLA)\u5927\u6a21\u578b\u7684\u53d1\u5c55\u53d6\u5f97\u91cd\u8981\u7a81\u7834\u3002\u5f53\u524d\u9762\u4e34\u7684\u6838\u5fc3\u79d1\u5b66\u95ee\u9898\u5305\u62ec\uff1a\u5982\u4f55\u63d0\u9ad8\u6570\u636e\u6548\u7387\u5e76\u89e3\u51b3\u771f\u5b9e\u573a\u666f\u6570\u636e\u83b7\u53d6\u56f0\u96be\u3001\u5982\u4f55\u589e\u5f3a\u6a21\u578b\u5728\u65b0\u573a\u666f\u65b0\u4efb\u52a1\u4e2d\u7684\u6cdb\u5316\u80fd\u529b\u3001\u5982\u4f55\u5b9e\u73b0\u6709\u6548\u7684\u4eff\u771f\u5230\u73b0\u5b9e\u8fc1\u79fb\uff0c\u4ee5\u53ca\u5982\u4f55\u5c06VLA\u5927\u6a21\u578b\u7684\u5f3a\u5927\u8868\u5f81\u80fd\u529b\u4e0e\u5b9e\u65f6\u673a\u5668\u4eba\u63a7\u5236\u9700\u6c42\u76f8\u5e73\u8861\u3002\u8fd9\u4e9b\u6311\u6218\u63a8\u52a8\u7740\u7aef\u5230\u7aef\u5b66\u4e60\u65b9\u6cd5\u4e0e\u5927\u6a21\u578b\u7684\u6df1\u5ea6\u878d\u5408\uff0c\u901a\u8fc7\u591a\u6a21\u6001\u4fe1\u606f\u7684\u6709\u6548\u6574\u5408\u6765\u63d0\u5347\u7cfb\u7edf\u6027\u80fd\u3002<\/span><\/p>\n


    <\/span><\/h4>\n

    VLA\u5927\u6a21\u578b\uff08Vision-Language-Action Model\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    \u4e0d\u540c\u4e8e\u7aef\u5230\u7aef\u5c0f\u6a21\u578b\uff08\u5982\u4e13\u95e8\u7528\u4e8e\u67d0\u9879\u4efb\u52a1\u7684\u6a21\u4eff\u5b66\u4e60\u6a21\u578b\u6216RL policy\uff09\uff0cVLA\u5927\u6a21\u578b\u5177\u5907\u4eba\u7c7b\u5e38\u8bc6\uff0c\u80fd\u591f\u63a5\u53d7\u5f00\u653e\u6307\u4ee4\u3001\u7406\u89e3\u5f00\u653e\u573a\u666f\uff1b\u4e0d\u540c\u4e8e\u8bf8\u591a\u5c06\u975e\u5177\u8eabLLM\/VLM\u4e0e\u673a\u5668\u4eba\u7cfb\u7edf\u4ee5\u642d\u79ef\u6728\u822c\u7ec4\u5408\u7684\u5de5\u4f5c\uff0cVLA\u5927\u6a21\u578b\u9700\u8981\u6574\u4f53\u8bad\u7ec3\u3002\u8fd9\u90e8\u5206\u5de5\u4f5c\u95e8\u69db\u6700\u9ad8\u3001\u96be\u5ea6\u6700\u5927\uff0c\u4f46\u6b63\u662fVLA\u65b9\u9762\u7684\u7814\u7a76\u8fdb\u5c55\u548c\u524d\u666f\uff0c\u63a8\u52a8\u4e86\u5177\u8eab\u667a\u80fd\u7684\u6295\u8d44\u70ed\u6f6e\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    1. \u7528\u4e8e\u5bfc\u822a\u7684VLA\u5927\u6a21\u578b\uff1aNaVid\uff0cNaVILA<\/span><\/p>\n


    <\/section>\n

    Zhang, Jiazhao, et al. “Navid: Video-based vlm plans the next step for vision-and-language navigation.” arXiv preprint arXiv:2402.15852<\/em> (2024). https:\/\/arxiv.org\/pdf\/2402.15852<\/span><\/p>\n

    Cheng, An-Chieh, et al. “NaVILA: Legged Robot Vision-Language-Action Model for Navigation.” arXiv preprint arXiv:2412.04453<\/em> (2024). https:\/\/arxiv.org\/abs\/2412.04453<\/span><\/p>\n


    <\/span><\/p>\n

    2. \u7528\u4e8e\u64cd\u7eb5\u7684VLA\u5927\u6a21\u578b\uff1aOpenVLA\uff0c\u03c00<\/span><\/p>\n


    <\/section>\n

    Kim, Moo Jin, et al. “OpenVLA: An Open-Source Vision-Language-Action Model.” arXiv preprint arXiv:2406.09246<\/em> (2024). https:\/\/arxiv.org\/abs\/2406.09246<\/span><\/p>\n

    \u03c00: Our First Generalist Policy https:\/\/www.physicalintelligence.company\/blog\/pi0<\/span><\/p>\n


    <\/span><\/h3>\n

    \u5b66\u4e60\u8303\u5f0f<\/span><\/strong><\/span><\/h3>\n


    <\/span><\/h4>\n

    \u4ece\u771f\u5b9e\u6570\u636e\u4e2d\u5b66\u4e60\uff08learning from real data\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    Learning from Real Data\u662f\u6307\u673a\u5668\u4eba\u901a\u8fc7\u76f4\u63a5\u4ece\u771f\u5b9e\u4e16\u754c\u73af\u5883\u4e2d\u6536\u96c6\u7684\u6570\u636e\u8fdb\u884c\u5b66\u4e60\uff0c\u8fd9\u4e9b\u6570\u636e\u5305\u62ec\u89c6\u89c9\u3001\u89e6\u89c9\u3001\u52a8\u4f5c\u7b49\u591a\u6a21\u6001\u4fe1\u606f\u3002\u8fd1\u5e74\u6765\uff0c\u968f\u7740RT-1\u3001RT-2\u7b49\u5927\u6a21\u578b\u7684\u53d1\u5c55\uff0c\u8fd9\u4e00\u8303\u5f0f\u8d8a\u6765\u8d8a\u53d7\u5230\u91cd\u89c6\u3002\u5728\u8fd9\u4e2a\u65b9\u5411\u4e0b\uff0c\u9762\u4e34\u7684\u6838\u5fc3\u6311\u6218\u5305\u62ec\uff1a\u6570\u636e\u6536\u96c6\u7684\u9ad8\u6210\u672c\u548c\u4f4e\u6548\u7387\u95ee\u9898\uff08\u9700\u8981\u6602\u8d35\u7684\u8bbe\u5907\u548c\u5927\u91cf\u4eba\u529b\uff09\u3001\u6570\u636e\u8d28\u91cf\u548c\u591a\u6837\u6027\u7684\u4fdd\u8bc1\uff08\u5982\u4f55\u786e\u4fdd\u6536\u96c6\u7684\u6570\u636e\u5177\u6709\u8db3\u591f\u7684\u4ee3\u8868\u6027\u548c\u8986\u76d6\u6027\uff09\u3001\u8de8\u5e73\u53f0\u6570\u636e\u7684\u6807\u51c6\u5316\u95ee\u9898\uff08\u4e0d\u540c\u673a\u5668\u4eba\u5e73\u53f0\u7684\u6570\u636e\u683c\u5f0f\u4e0d\u7edf\u4e00\uff09\u3001\u4ee5\u53ca\u5982\u4f55\u6709\u6548\u6574\u5408\u548c\u5229\u7528\u4eba\u7c7b\u793a\u8303\u6570\u636e\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Brohan, Albert, et al. “RT-1: Robotics Transformer for Real-world Control at Scale.” arXiv preprint arXiv:2212.06817 (2022).https:\/\/arxiv.org\/abs\/2212.06817<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u9996\u6b21\u5c55\u793a\u4e86\u57fa\u4e8e\u5927\u89c4\u6a21\u771f\u5b9e\u6570\u636e\u8bad\u7ec3\u7684\u673a\u5668\u4ebatransformer\u6a21\u578b\uff0c\u5efa\u7acb\u4e86\u4ece\u89c6\u89c9\u5230\u63a7\u5236\u7684\u7aef\u5230\u7aef\u6620\u5c04\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Majumdar, Arjun, et al. “Openeqa: Embodied question answering in the era of foundation models.” Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition<\/em>. 2024.<\/span><\/p>\n

    https:\/\/openaccess.thecvf.com\/content\/CVPR2024\/html\/Majumdar_OpenEQA_Embodied_Question_Answering_in_the_Era_of_Foundation_Models_CVPR_2024_paper.html<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u5c55\u793a\u4e86\u5982\u4f55\u5229\u7528\u771f\u5b9e\u573a\u666f\u6570\u636e\u6765\u8bad\u7ec3\u5f00\u653e\u8bcd\u6c47\u7684\u5177\u8eab\u95ee\u7b54\u7cfb\u7edf\uff0c\u4f53\u73b0\u4e86\u771f\u5b9e\u6570\u636e\u5bf9\u8bed\u8a00\u5f15\u5bfc\u4efb\u52a1\u7684\u91cd\u8981\u6027\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Brohan, Albert, et al. “RT-2: Vision-language-action Models Transfer Web Knowledge to Robotic Control.” Conference on Robot Learning. PMLR (2023): 2165-2183.https:\/\/arxiv.org\/abs\/2307.15818<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u901a\u8fc7\u7ed3\u5408\u4e92\u8054\u7f51\u9884\u8bad\u7ec3\u548c\u771f\u5b9e\u673a\u5668\u4eba\u6570\u636e\u5fae\u8c03\uff0c\u5b9e\u73b0\u4e86\u77e5\u8bc6\u8fc1\u79fb\u548c\u6280\u80fd\u6cdb\u5316\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u4ece\u4eff\u771f\u5230\u73b0\u5b9e\uff08sim2real \/ real2sim2real \uff09<\/span><\/strong><\/span><\/h4>\n


    <\/p>\n

    Sim2Real\/Real2Sim2Real\u662f\u5177\u8eab\u667a\u80fd\u9886\u57df\u4e2d\u5b9e\u73b0\u865a\u62df\u73af\u5883\u4e0e\u7269\u7406\u4e16\u754c\u5bf9\u9f50\u7684\u5173\u952e\u6280\u672f\u8303\u5f0f\uff0c\u65e8\u5728\u5c06\u4eff\u771f\u73af\u5883\u4e2d\u8bad\u7ec3\u7684\u6a21\u578b\u80fd\u529b\u6709\u6548\u8fc1\u79fb\u5230\u771f\u5b9e\u4e16\u754c\u3002\u5176\u4e2dReal2Sim2Real\u901a\u8fc7\u626b\u63cf\u548c\u91cd\u5efa\u771f\u5b9e\u573a\u666f\u6784\u5efa\u201c\u6570\u5b57\u5b6a\u751f\u201d\u73af\u5883\uff0c\u5728\u4eff\u771f\u4e2d\u8fdb\u884c\u5f3a\u5316\u5b66\u4e60\u8bad\u7ec3\u540e\u518d\u8fc1\u79fb\u56de\u771f\u5b9e\u4e16\u754c\u3002\u5728\u8fd9\u4e2a\u65b9\u5411\u4e0b\uff0c\u5982\u4f55\u63d0\u5347\u4eff\u771f\u73af\u5883\u7684\u7269\u7406\u771f\u5b9e\u6027\u4ee5\u7f29\u5c0f\u4e0e\u73b0\u5b9e\u4e16\u754c\u7684\u5dee\u8ddd\uff0c\u5982\u4f55\u786e\u4fdd\u6a21\u578b\u5728\u8fc1\u79fb\u8fc7\u7a0b\u4e2d\u7684\u9c81\u68d2\u6027\u548c\u6cdb\u5316\u80fd\u529b\uff0c\u4ee5\u53ca\u5982\u4f55\u9ad8\u6548\u6536\u96c6\u548c\u5229\u7528\u771f\u5b9e\u4e16\u754c\u7684\u6570\u636e\u6765\u4f18\u5316\u8fc1\u79fb\u6548\u679c\u662f\u6700\u91cd\u8981\u7684\u6311\u6218\u4e4b\u4e00\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u4e00\u5f00\u59cb\u7684\u63a8\u8350\u7efc\u8ff0\u6587\u7ae0\uff0c\u5176\u4e2d\u6709\u7cfb\u7edf\u7684\u68b3\u7406Sim2Real\/Real2Sim2Real\u76f8\u5173\u7684\u6a21\u5757\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u4ece\u4e16\u754c\u6a21\u578b\u4e2d\u5b66\u4e60\uff08learn from world model\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    \u4ece\u4e16\u754c\u6a21\u578b\u4e2d\u5b66\u4e60\u662f\u6307\u901a\u8fc7\u6784\u5efa\u548c\u5229\u7528\u4e16\u754c\u6a21\u578b\uff08World Model\uff09\uff0c\u8ba9\u673a\u5668\u4eba\u6216\u5177\u8eab\u667a\u80fd\u4f53\u5728\u865a\u62df\u6216\u7269\u7406\u73af\u5883\u4e2d\u5b66\u4e60\u548c\u63a8\u7406\u3002\u4e16\u754c\u6a21\u578b\u901a\u8fc7\u6a21\u62df\u73af\u5883\u7684\u7269\u7406\u89c4\u5f8b\u548c\u56e0\u679c\u5173\u7cfb\uff0c\u9884\u6d4b\u8f93\u5165\u4e0e\u8f93\u51fa\u4e4b\u95f4\u7684\u72b6\u6001\u8f6c\u53d8\uff0c\u4ece\u800c\u4e3a\u673a\u5668\u4eba\u63d0\u4f9b\u611f\u77e5\u3001\u89c4\u5212\u548c\u51b3\u7b56\u7684\u80fd\u529b\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u7136\u800c\uff0c\u5f53\u524dWorld Model\u4ecd\u9762\u4e34\u51e0\u4e2a\u5173\u952e\u6311\u6218\uff1a\u5982\u4f55\u63d0\u5347\u6a21\u578b\u7684\u6cdb\u5316\u80fd\u529b\u4ee5\u9002\u5e94\u672a\u89c1\u573a\u666f\uff0c\u5982\u4f55\u6709\u6548\u5b66\u4e60\u548c\u8868\u8fbe\u7269\u7406\u89c4\u5f8b\u4e0e\u56e0\u679c\u5173\u7cfb\uff0c\u5982\u4f55\u5e73\u8861\u6570\u636e\u6548\u7387\u4e0e\u8ba1\u7b97\u6210\u672c\uff0c\u4ee5\u53ca\u5982\u4f55\u786e\u4fdd\u5728\u5b9e\u65f6\u4efb\u52a1\u4e2d\u7684\u9c81\u68d2\u6027\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Liu, Yang, et al. “Aligning cyber space with physical world: A comprehensive survey on embodied ai.” arXiv preprint arXiv:2407.06886<\/em> (2024).https:\/\/arxiv.org\/abs\/2407.06886<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u4e00\u5f00\u59cb\u7684\u63a8\u8350\u7efc\u8ff0\u6587\u7ae0\uff0c\u5176\u4e2d\u6709\u7cfb\u7edf\u7684\u68b3\u7406World Model\u76f8\u5173\u7684\u6a21\u5757\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u8de8\u5177\u8eab\u5b66\u4e60\uff08cross-embodiment learning\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/p>\n

    \u8de8\u5177\u8eab\u5b66\u4e60\u662f\u6307\u901a\u8fc7\u5171\u4eab\u548c\u8fc1\u79fb\u77e5\u8bc6\uff0c\u4f7f\u673a\u5668\u4eba\u80fd\u591f\u5728\u4e0d\u540c\u5f62\u6001\uff08\u5982\u5355\u81c2\u3001\u53cc\u81c2\u3001\u4eba\u5f62\u3001\u8f6e\u5f0f\u7b49\uff09\u548c\u5e73\u53f0\u4e4b\u95f4\u8fdb\u884c\u5b66\u4e60\u548c\u9002\u5e94\uff0c\u4ece\u800c\u63d0\u5347\u5176\u5728\u591a\u6837\u5316\u4efb\u52a1\u4e2d\u7684\u6cdb\u5316\u80fd\u529b\u548c\u534f\u4f5c\u80fd\u529b\u3002\u8fd9\u4e00\u65b9\u5411\u9762\u4e34\u7684\u4e3b\u8981\u79d1\u5b66\u95ee\u9898\u5305\u62ec\uff1a\u5982\u4f55\u8bbe\u8ba1\u7edf\u4e00\u7684\u52a8\u4f5c\u548c\u611f\u77e5\u8868\u793a\u4ee5\u8de8\u8d8a\u4e0d\u540c\u673a\u5668\u4eba\u5f62\u6001\u7684\u5dee\u5f02\u3001\u5982\u4f55\u5904\u7406\u591a\u6a21\u6001\u6570\u636e\u7684\u6807\u51c6\u5316\u4e0e\u6574\u5408\uff0c\u4ee5\u53ca\u5982\u4f55\u6709\u6548\u5730\u89e3\u51b3\u6a21\u62df\u4e0e\u73b0\u5b9e\u4e4b\u95f4\u7684\u5dee\u8ddd\uff08sim-to-real gap\uff09\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Wang, Zhiqiang, et al. “All robots in one: A new standard and unified dataset for versatile, general-purpose embodied agents.” arXiv preprint arXiv:2408.10899<\/em> (2024).https:\/\/arxiv.org\/abs\/2408.10899<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u7cfb\u7edf\u603b\u7ed3\u4e86\u8de8\u5e73\u53f0\u548c\u591a\u5f62\u6001\u673a\u5668\u4eba\u5b66\u4e60\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u63d0\u51fa\u4e86\u7edf\u4e00\u6570\u636e\u683c\u5f0f\u548c\u591a\u6a21\u6001\u6574\u5408\u7684\u89e3\u51b3\u65b9\u6848\uff0c\u662f\u5177\u8eab\u667a\u80fd\u9886\u57df cross-embodiment learning \u7814\u7a76\u7684\u91cd\u8981\u53c2\u8003\u6587\u732e\u3002<\/span><\/p>\n


    <\/p>\n

    <\/p>\n


    <\/h4>\n

    \u4ece\u5176\u4ed6\u9886\u57df\u77e5\u8bc6\u4e2d\u5b66\u4e60\uff08learning from other domain knowledge\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    \u4ece\u5176\u4ed6\u9886\u57df\u77e5\u8bc6\u4e2d\u5b66\u4e60\u662f\u6307\u901a\u8fc7\u5229\u7528\u5176\u4ed6\u9886\u57df\uff08\u5982\u4e92\u8054\u7f51\u89c6\u9891\u3001\u81ea\u7136\u8bed\u8a00\u5904\u7406\u6216\u8ba1\u7b97\u673a\u89c6\u89c9\uff09\u4e2d\u5df2\u6709\u7684\u4e30\u5bcc\u6570\u636e\u548c\u6a21\u578b\uff0c\u63d0\u53d6\u901a\u7528\u77e5\u8bc6\u5e76\u5c06\u5176\u8fc1\u79fb\u5230\u673a\u5668\u4eba\u5b66\u4e60\u4e2d\uff0c\u4ee5\u5f25\u8865\u673a\u5668\u4eba\u9886\u57df\u6570\u636e\u7a00\u7f3a\u548c\u4efb\u52a1\u590d\u6742\u6027\u5e26\u6765\u7684\u6311\u6218\u3002\u8fd9\u4e00\u65b9\u5411\u7684\u6838\u5fc3\u79d1\u5b66\u95ee\u9898\u5305\u62ec\u5982\u4f55\u4ece\u5f02\u6784\u6570\u636e\u4e2d\u63d0\u53d6\u4e0e\u673a\u5668\u4eba\u4efb\u52a1\u76f8\u5173\u7684\u77e5\u8bc6\uff0c\u4ee5\u53ca\u5982\u4f55\u5e94\u5bf9\u6570\u636e\u5206\u5e03\u5dee\u5f02\u3001\u7f3a\u4e4f\u4f4e\u7ea7\u611f\u77e5\u4fe1\u606f\uff08\u5982\u529b\u89c9\u548c\u89e6\u89c9\uff09\u548c\u9ad8\u7ef4\u6570\u636e\u5904\u7406\u7684\u6311\u6218\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    McCarthy, Robert, et al. “Towards Generalist Robot Learning from Internet Video: A Survey.” arXiv preprint arXiv:2404.19664<\/em> (2024).https:\/\/arxiv.org\/abs\/2404.19664<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u7cfb\u7edf\u7efc\u8ff0\u4e86\u5982\u4f55\u4ece\u4e92\u8054\u7f51\u89c6\u9891\u7b49\u5176\u4ed6\u9886\u57df\u77e5\u8bc6\u4e2d\u63d0\u53d6\u548c\u8fc1\u79fb\u4fe1\u606f\u4ee5\u63d0\u5347\u673a\u5668\u4eba\u5b66\u4e60\u80fd\u529b\uff0c\u662f\u5177\u8eab\u667a\u80fd\u9886\u57df\u5b66\u4e60\u8de8\u9886\u57df\u77e5\u8bc6\u7684\u91cd\u8981\u53c2\u8003\u6587\u732e\u3002<\/span><\/p>\n


    <\/p>\n

    <\/p>\n


    <\/p>\n

    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n<\/section>\n
    \n
    \u4efb\u52a1\/\u573a\u666f<\/span><\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

    <\/p>\n


    <\/span><\/p>\n

    \u8fd9\u4e2a\u6a21\u5757\uff0c\u662f\u5bf9\u524d\u9762\u4e24\u4e2a\u4e3b\u9898\u7684\u4e00\u4e2a\u7efc\u5408\u8ba8\u8bba\uff0c\u6700\u7ec8\uff0c\u5177\u8eab\u667a\u80fd\u7684\u4ef7\u503c\u8981\u5728\u5b9e\u9645\u5e94\u7528\u4efb\u52a1\u4e2d\u5f97\u5230\u9a8c\u8bc1\u3002\u8fd0\u52a8\u63a7\u5236\u3001\u7269\u4f53\u64cd\u4f5c\u3001\u8fdc\u7a0b\u64cd\u4f5c\u3001\u573a\u666f\u7406\u89e3\u4e0e\u4ea4\u4e92\u7b49\u65b9\u5411\u5404\u6709\u5176\u72ec\u7279\u7684\u6311\u6218\u3002\u4f8b\u5982\uff1a\u5982\u4f55\u5b9e\u73b0\u50cf\u4eba\u7c7b\u4e00\u6837\u81ea\u7136\u6d41\u7545\u7684\u53cc\u8db3\u884c\u8d70\uff1f\u5982\u4f55\u8ba9\u673a\u5668\u4eba\u80fd\u591f\u7406\u89e3\u5e76\u4f7f\u7528\u5404\u79cd\u65e5\u5e38\u5de5\u5177\uff1f\u5982\u4f55\u5728\u4eba\u673a\u534f\u4f5c\u4e2d\u5b9e\u73b0\u9ad8\u6548\u7684\u4ea4\u4e92\uff1f\u8fd9\u4e9b\u95ee\u9898\u4e0d\u4ec5\u9700\u8981\u89e3\u51b3\u6280\u672f\u6311\u6218\uff0c\u8fd8\u9700\u8981\u601d\u8003\u5982\u4f55\u5c06\u5177\u8eab\u667a\u80fd\u66f4\u597d\u5730\u670d\u52a1\u4e8e\u5b9e\u9645\u5e94\u7528\u573a\u666f\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u79fb\u52a8\uff08locomotion\uff09<\/span><\/strong><\/span><\/h3>\n


    <\/p>\n

    \u79fb\u52a8\u5728\u673a\u5668\u4eba\u9886\u57df\u901a\u5e38\u6307\u673a\u5668\u4eba\u5728\u73af\u5883\u4e2d\u8fdb\u884c\u81ea\u4e3b\u79fb\u52a8\u548c\u5bfc\u822a\u7684\u80fd\u529b\uff0c\u5305\u62ec\u901a\u8fc7\u8f6e\u5f0f\u3001\u5c65\u5e26\u5f0f\u3001\u8db3\u5f0f\uff08\u5982\u56db\u8db3\u3001\u4eff\u751f\u7b49\uff09\u4ee5\u53ca\u5176\u4ed6\u751f\u7269\u542f\u53d1\u5f0f\u65b9\u5f0f\u6765\u5b8c\u6210\u8d8a\u969c\u3001\u4fdd\u6301\u5e73\u8861\u548c\u6267\u884c\u79fb\u52a8\u4efb\u52a1\u3002\u5728\u8fd9\u4e00\u65b9\u5411\u4e0b\u9762\u4e34\u7684\u4e3b\u8981\u79d1\u5b66\u95ee\u9898\u4e3b\u8981\u662f\uff1a\u5982\u4f55\u5b9e\u73b0\u9ad8\u901f\u3001\u9ad8\u52a8\u6001\u6761\u4ef6\u4e0b\u7684\u7a33\u5b9a\u63a7\u5236\uff1f<\/span><\/p>\n


    <\/span><\/p>\n

    Liu, Min, et al. “Visual Whole-body Control for Legged Loco-manipulation.” arXiv preprint arXiv:2402.16967 (2024).<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u63d0\u51fa\u57fa\u4e8e\u89c6\u89c9\u7684\u5168\u8eab\u63a7\u5236\u6846\u67b6\uff0c\u5b9e\u73b0\u4e86\u817f\u90e8\u8fd0\u52a8\u548c\u673a\u68b0\u81c2\u64cd\u4f5c\u7684\u534f\u8c03\u63a7\u5236\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Bledt, Gabriel, et al. “MIT Cheetah 3: Design and Control of a Robust, Dynamic Quadruped Robot.” IEEE\/RSJ International Conference on Intelligent Robots and Systems (2018): 2245-2252.<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u5c55\u793a\u4e86\u56db\u8db3\u673a\u5668\u4eba\u5728\u9c81\u68d2\u6027\u548c\u52a8\u6001\u6027\u80fd\u4e0a\u7684\u7a81\u7834\uff0c\u5305\u62ec\u8df3\u8dc3\u548c\u8d8a\u969c\u7b49\u9ad8\u96be\u5ea6\u52a8\u4f5c\u7684\u5b9e\u73b0\u3002<\/span><\/p>\n


    <\/span><\/h3>\n

    \u64cd\u63a7\uff08Manipulation\uff09<\/span><\/strong><\/span><\/h3>\n


    <\/p>\n

    \u64cd\u63a7\u662f\u6307\u673a\u5668\u4eba\u901a\u8fc7\u611f\u77e5\u3001\u89c4\u5212\u548c\u6267\u884c\u4e00\u7cfb\u5217\u52a8\u4f5c\uff0c\u4e0e\u73af\u5883\u4e2d\u7684\u7269\u4f53\u8fdb\u884c\u4ea4\u4e92\u7684\u8fc7\u7a0b\uff0c\u5305\u62ec\u6293\u53d6\u3001\u79fb\u52a8\u3001\u8c03\u6574\u548c\u4f7f\u7528\u7269\u4f53\u7b49\u4efb\u52a1\u3002\u5176\u76ee\u6807\u662f\u4f7f\u673a\u5668\u4eba\u80fd\u591f\u5728\u52a8\u6001\u548c\u590d\u6742\u7684\u73af\u5883\u4e2d\u5b8c\u6210\u591a\u6837\u5316\u7684\u4efb\u52a1\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u5de5\u4e1a\u3001\u670d\u52a1\u548c\u533b\u7597\u7b49\u9886\u57df\u3002\u64cd\u63a7\u9762\u4e34\u7684\u4e3b\u8981\u79d1\u5b66\u95ee\u9898\u5305\u62ec\uff1a\u5982\u4f55\u5728\u975e\u7ed3\u6784\u5316\u73af\u5883\u4e2d\u5b9e\u73b0\u9ad8\u7cbe\u5ea6\u7684\u611f\u77e5\u4e0e\u51b3\u7b56\u3001\u5982\u4f55\u751f\u6210\u7075\u6d3b\u4e14\u9ad8\u6548\u7684\u64cd\u4f5c\u7b56\u7565\uff0c\u4ee5\u53ca\u5982\u4f55\u5e94\u5bf9\u5b9e\u65f6\u6027\u8981\u6c42\u3001\u73af\u5883\u4e0d\u786e\u5b9a\u6027\u548c\u591a\u6a21\u6001\u6570\u636e\u878d\u5408\u7684\u6311\u6218\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Zheng, Ying, et al. “A Survey of Embodied Learning for Object-Centric Robotic Manipulation.” arXiv preprint arXiv:2408.11537<\/em> (2024).https:\/\/arxiv.org\/abs\/2408.11537<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u5728\u8fd9\u4e2a\u7efc\u8ff0\u91cc\u9762\uff0c\u7cfb\u7edf\u7efc\u8ff0\u4e86\u4ee5\u7269\u4f53\u4e3a\u4e2d\u5fc3\u7684\u673a\u5668\u4eba\u64cd\u63a7\u9886\u57df\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u6db5\u76d6\u611f\u77e5\u3001\u7b56\u7565\u751f\u6210\u548c\u4efb\u52a1\u5bfc\u5411\u5b66\u4e60\u7b49\u5173\u952e\u65b9\u5411\uff0c\u662f\u7814\u7a76\u5177\u8eab\u667a\u80fd\u9886\u57df\u64cd\u63a7\u6280\u672f\u7684\u91cd\u8981\u53c2\u8003\u6587\u732e\u3002<\/span><\/p>\n


    <\/section>\n

    <\/span><\/p>\n

    <\/p>\n


    <\/span><\/h4>\n

    \u6293\u53d6\uff08Grasping\uff09<\/span><\/strong><\/span><\/h4>\n


    <\/span><\/p>\n

    \u6293\u53d6\u4f5c\u4e3a\u6700\u57fa\u7840\u7684\u673a\u5668\u4eba\u64cd\u7eb5\u4efb\u52a1\uff0c\u53ef\u80fd\u4e5f\u662f\u6700\u5148\u88ab\u7814\u7a76\u5230\u6210\u719f\u7684\u4efb\u52a1\uff0c\u503c\u5f97\u4e13\u95e8\u91cd\u70b9\u8ba8\u8bba\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    GraspNet\uff1ahttps:\/\/graspnet.net\/<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aGraspNet \u662f\u4e00\u4e2a\u5927\u89c4\u6a21\u3001\u591a\u6837\u5316\u3001\u652f\u6301\u6a21\u62df\u4e0e\u771f\u5b9e\u8bc4\u4f30\u7684\u673a\u5668\u4eba\u6293\u53d6\u6570\u636e\u96c6\u548c\u57fa\u51c6\u5e73\u53f0\uff0c\u4e3a\u5177\u8eab\u667a\u80fd\u6293\u53d6\u4efb\u52a1\u63d0\u4f9b\u4e86\u6807\u51c6\u5316\u8bc4\u4f30\u548c\u5f3a\u5927\u7684\u5de5\u5177\u652f\u6301\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Shi, Jun, et al. “Asgrasp: Generalizable transparent object reconstruction and 6-dof grasp detection from rgb-d active stereo camera.” 2024 IEEE International Conference on Robotics and Automation (ICRA)<\/em>. IEEE, 2024.https:\/\/pku-epic.github.io\/ASGrasp\/<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aASGrasp\u662f\u4e00\u79cd\u57fa\u4e8eRGB-D\u4e3b\u52a8\u7acb\u4f53\u76f8\u673a\u76846\u81ea\u7531\u5ea6\u6293\u53d6\u68c0\u6d4b\u7f51\u7edc\uff0c\u4e13\u4e3a\u89e3\u51b3\u900f\u660e\u548c\u955c\u9762\u7269\u4f53\u6293\u53d6\u95ee\u9898\u800c\u8bbe\u8ba1\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Wei, Songlin, et al. “D3<\/sup> RoMa: Disparity Diffusion-based Depth Sensing for Material-Agnostic Robotic Manipulation.” ECCV 2024 Workshop on Wild 3D: 3D Modeling, Reconstruction, and Generation in the Wild<\/em>. 2024.https:\/\/openreview.net\/forum?id=jFRUsiuhUy<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aD3RoMa \u63d0\u51fa\u4e86\u4e00\u4e2a\u57fa\u4e8e\u6269\u6563\u6a21\u578b\u7684\u6df1\u5ea6\u4f30\u8ba1\u6846\u67b6\uff0c\u80fd\u591f\u5728\u590d\u6742\u573a\u666f\u4e2d\u751f\u6210\u9ad8\u8d28\u91cf\u7684\u6df1\u5ea6\u56fe\uff0c\u7279\u522b\u662f\u5728\u900f\u660e\u548c\u955c\u9762\u7269\u4f53\u7b49\u4f20\u7edf\u65b9\u6cd5\u5931\u6548\u7684\u573a\u666f\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Wang, Ruicheng, et al. “Dexgraspnet: A large-scale robotic dexterous grasp dataset for general objects based on simulation.” 2023 IEEE International Conference on Robotics and Automation (ICRA)<\/em>. IEEE, 2023.https:\/\/ieeexplore.ieee.org\/abstract\/document\/10160982\/<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aDexGraspNet \u63d0\u51fa\u4e86\u4e00\u4e2a\u5927\u89c4\u6a21\u7075\u5de7\u6293\u53d6\u6570\u636e\u96c6\uff0c\u4e13\u6ce8\u4e8e\u7075\u5de7\u624b\u5728\u590d\u6742\u573a\u666f\u4e2d\u7684\u6293\u53d6\u4efb\u52a1\uff0c\u6db5\u76d6\u4e86\u591a\u6837\u5316\u7684\u7269\u4f53\u548c\u6293\u53d6\u7b56\u7565\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Zhang, Jialiang, et al. “DexGraspNet 2.0: Learning Generative Dexterous Grasping in Large-scale Synthetic Cluttered Scenes.” 8th Annual Conference on Robot Learning<\/em>. 2024.https:\/\/openreview.net\/forum?id=V6XHKFicwc<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1aDexGraspNet 2.0 \u63d0\u51fa\u4e86\u4e00\u4e2a\u5927\u89c4\u6a21\u5408\u6210\u6293\u53d6\u6570\u636e\u96c6\uff0c\u5e76\u7ed3\u5408\u751f\u6210\u5f0f\u7075\u5de7\u6293\u53d6\u7b97\u6cd5\u5728\u590d\u6742\u573a\u666f\u4e2d\u5b9e\u73b0\u9ad8\u6210\u529f\u7387\u7684\u7075\u6d3b\u64cd\u4f5c\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    \u9065\u64cd\u4f5c\uff08Teleoperation \uff09<\/span><\/strong><\/span><\/h3>\n

    <\/span><\/section>\n

    \u9065\u64cd\u4f5c\u662f\u5177\u8eab\u667a\u80fd\u9886\u57df\u8fde\u63a5\u4eba\u7c7b\u667a\u80fd\u4e0e\u673a\u5668\u4eba\u7cfb\u7edf\u7684\u5173\u952e\u6865\u6881\u3002\u5b83\u901a\u8fc7\u4eba\u7c7b\u64cd\u4f5c\u8005\u7684\u8fdc\u7a0b\u63a7\u5236\uff0c\u4e3a\u673a\u5668\u4eba\u7cfb\u7edf\u63d0\u4f9b\u4e86\u83b7\u53d6\u548c\u5b66\u4e60\u590d\u6742\u64cd\u4f5c\u6280\u80fd\u7684\u91cd\u8981\u9014\u5f84\u3002\u5f53\u524d\u9065\u64cd\u4f5c\u7814\u7a76\u7684\u6838\u5fc3\u79d1\u5b66\u95ee\u9898\u662f\u5982\u4f55\u5efa\u7acb\u9ad8\u6548\u81ea\u7136\u7684\u4eba\u673a\u6280\u80fd\u6620\u5c04\u673a\u5236\uff0c\u8fd9\u6d89\u53ca\u64cd\u4f5c\u6280\u80fd\u7684\u672c\u8d28\u8868\u5f81\u3001\u4eba\u673a\u4ea4\u4e92\u7684\u81ea\u7136\u6027\u4ee5\u53ca\u63a7\u5236\u7cfb\u7edf\u7684\u5b9e\u65f6\u6027\u7b49\u591a\u4e2a\u65b9\u9762\u7684\u6311\u6218\u3002<\/span><\/p>\n


    <\/span><\/p>\n

    Darvish, Kourosh, et al. “Teleoperation of humanoid robots: A survey.” IEEE Transactions on Robotics<\/em> 39.3 (2023): 1706-1727.https:\/\/ieeexplore.ieee.org\/abstract\/document\/10035484<\/span><\/p>\n

    \u63a8\u8350\u7406\u7531\uff1a\u8fd9\u7bc7\u6587\u732e\u5168\u9762\u7efc\u8ff0\u4e86\u4eba\u5f62\u673a\u5668\u4eba\u9065\u64cd\u4f5c\u9886\u57df\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u6db5\u76d6\u6280\u672f\u6311\u6218\u3001\u63a7\u5236\u65b9\u6cd5\u3001\u754c\u9762\u8bbe\u8ba1\u53ca\u5e94\u7528\u573a\u666f\uff0c\u7279\u522b\u5f3a\u8c03\u4e86\u9065\u64cd\u4f5c\u5728\u590d\u6742\u73af\u5883\u4e2d\u7684\u6f5c\u529b\u4e0e\u672a\u6765\u53d1\u5c55\u65b9\u5411\uff0c\u662f\u7814\u7a76\u5177\u8eab\u667a\u80fd\u4e0e\u9065\u64cd\u4f5c\u6280\u672f\u7684\u91cd\u8981\u53c2\u8003\u3002<\/span><\/p>\n


    <\/p>\n

    <\/p>\n


    <\/span><\/h2>\n


    <\/span><\/p>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    \u793e\u533a\u4f19\u4f34\u81f4\u8c22<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/p>\n

    \u611f\u8c22\u4ee5\u4e0b\u793e\u533a\u4f19\u4f34\u7684\u652f\u6301\u3002<\/span><\/p>\n


    <\/span><\/h2>\n

    <\/h2>\n

    Datawhale<\/span><\/strong><\/span><\/h2>\n

    Datawhale\u6210\u7acb\u4e8e2018\u5e74\u2014\u2014\u662f\u56fd\u5185AI\u9886\u57df\u6700\u5927\u7684\u5f00\u6e90\u5b66\u4e60\u7ec4\u7ec7\uff0c\u6c47\u805a\u4e86\u4f17\u591a\u6709\u5f00\u6e90\u7cbe\u795e\u548c\u63a2\u7d22\u7cbe\u795e\u7684\u5f00\u6e90\u8d21\u732e\u8005\uff0c\u81f4\u529b\u4e8e\u6784\u5efa\u5f00\u6e90\u5b66\u4e60\u793e\u533a\uff0c\u548c\u5b66\u4e60\u8005\u4e00\u8d77\u6210\u957f\u3002\u6210\u7acb5\u5e74\u6765\uff0cDatawhale\u5f00\u6e90\u5b66\u4e60\u6210\u4e3a\u5168\u56fd\u77e5\u540d\u7684\u5b66\u4e60\u6d3b\u52a8\uff0c\u8986\u76d6\u4e86\u6d77\u5185\u59162000\u591a\u6240\u9662\u6821\uff0c\u81f4\u529b\u4e8e\u6253\u9020\u5173\u6ce8\u9752\u5e74\u4eba\u624d\u6210\u957f\u7684\u4e3b\u9635\u5730\uff0c\u52a0\u5feb\u9ad8\u6821\u548c\u4ea7\u4e1a\u7684\u94fe\u63a5\uff0c\u6301\u7eed\u63a8\u52a8\u9752\u5e74AI\u4eba\u624d\u57f9\u517b\u3002<\/span><\/h2>\n


    <\/span><\/h2>\n

    \u8ba1\u7b97\u673a\u89c6\u89c9life\u5e73\u53f0<\/span><\/strong><\/span><\/h2>\n

    \u8ba1\u7b97\u673a\u89c6\u89c9life (CV life) \u662f\u56fd\u5185\u9886\u5148\u7684\u673a\u5668\u4eba\u4e0e\u4eba\u5de5\u667a\u80fd\u9886\u57df\u7efc\u5408\u6027\u5e73\u53f0\uff0c\u7531\u4e2d\u56fd\u79d1\u5b66\u9662\u535a\u58eb\u7a0b\u5c0f\u516d\u521b\u7acb\u4e8e2018\u5e74\u3002\u4f5c\u4e3a\u673a\u5668\u4ebaAI\u9886\u57df\u7684\u7b2c\u4e00\u68af\u961f\u6280\u672f\u670d\u52a1\u5e73\u53f0\uff0cCV life\u81f4\u529b\u4e8e\u6253\u9020\u5b8c\u6574\u7684\u673a\u5668\u4ebaAI\u884c\u4e1a\u751f\u6001\u5708\uff0c\u6db5\u76d6\u5a92\u4f53\u3001\u6559\u80b2\u57f9\u8bad\u3001\u4eba\u624d\u670d\u52a1\u3001\u6280\u672f\u7814\u53d1\u3001\u5c31\u4e1a\u5bf9\u63a5\u7b49\u591a\u4e2a\u7ef4\u5ea6\u3002<\/span><\/p>\n


    <\/span><\/h2>\n

    \u5177\u8eab\u667a\u80fd\u77e5\u8bc6\u5e93\u793e\u533a<\/span><\/strong><\/span><\/h2>\n

    \u5b66\u4e60\u673a\u5668\u4eba\u4e0e\u5177\u8eab\u667a\u80fd\u6280\u672f\u524d\u6cbf\u6280\u672f\u5e73\u53f0\uff0c\u5305\u542b\u5177\u8eab\u667a\u80fd\u4ece\u6d45\u5165\u6df1\u7684\u77e5\u8bc6\u4f53\u7cfb\uff0c\u7acb\u5fd7\u4e0e\u5168\u7f51\u5177\u8eab\u667a\u80fd\u7814\u7a76\u4eba\u5458\u5171\u5efa\u6700\u5168\u5177\u8eab\u667a\u80fd\u77e5\u8bc6\u5e93\u548c\u6570\u636e\u5e73\u53f0\u3002<\/span><\/p>\n


    <\/p>\n

    <\/p>\n


    <\/section>\n

    <\/section>\n

    \u70b9\u51fb\u201c\u9605\u8bfb\u539f\u6587\u201d\uff0c\u62a5\u540d\u8bfb\u4e66\u4f1a<\/span><\/p>\n

    <\/mp-style-type><\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

    Jeffrey Fisher for Quanta Magazine \u5bfc\u8bed \u96c6\u667a\u4ff1\u4e50\u90e8\u8054\u5408\u4e0a\u6d77\u4ea4\u901a\u5927\u5b66\u52a9\u7406\u6559\u6388\u674e\u6c38\u9732\u3001\u94f6\u6cb3\u901a\u7528\u673a\u5668\u4eba\u5408\u4f19\u4eba\u53f2\u96ea\u677e\u3001\u5357\u4eac\u5927\u5b66LAMDA\u7ec4\u535a\u58eb\u751f\u9648\u96c4\u8f89\u3001\u9999\u6e2f\u5927\u5b66\u5728\u8bfb\u535a\u58eb\u751f\u7a46\u5c27\uff0c\u5171\u540c\u53d1\u8d77\u9996\u5b63\u300c\u5177\u8eab\u667a\u80fd\u300d\u8bfb\u4e66\u4f1a\u3002\u8bfb\u4e66\u4f1a\u8ba1\u5212\u91c7\u7528\u201c\u81ea\u4e0b\u800c\u4e0a\u201d\u7684\u5c42\u7ea7\u7ed3\u6784\uff0c\u63a2\u8ba8\u56db\u4e2a\u6838\u5fc3\u6a21\u5757\uff1a\u786c\u4ef6\u7cfb\u7edf…<\/p>\n","protected":false},"author":0,"featured_media":54523,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"special":[],"_links":{"self":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/54563"}],"collection":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=54563"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/54563\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/54523"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=54563"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=54563"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=54563"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=54563"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}