Hey everyone, I am one of the authors of the paper described in this article. We used reinforcement learning and recurrent neural networks to learn a controller which can ascend and descend stairs without any vision-based perception, meaning that it must rely entirely on proprioception to walk. This is the first time (to our knowledge) that a human-sized bipedal robot has been able to climb real-world stairs blind to the world, and we're pretty excited about the results of this research.<p>Here's a link to the Arxiv submission: <a href="https://arxiv.org/abs/2105.08328" rel="nofollow">https://arxiv.org/abs/2105.08328</a><p>And here's the accompanying video:
<a href="https://youtu.be/MPhEmC6b6XU" rel="nofollow">https://youtu.be/MPhEmC6b6XU</a><p>And an uninterrupted five-minute video of a test on an outdoor staircase:
<a href="https://youtu.be/nuhHiKEtaZQ" rel="nofollow">https://youtu.be/nuhHiKEtaZQ</a><p>Happy to answer any questions the HN crowd may have!