3 ms·
Classical control theory is often sufficient for dealing with changing loads, terrain and wear and tear. Rejecting unknown disturbances is what a control system
by Matti 5y ago
Classical control theory is often sufficient for dealing with changing loads, terrain and wear and tear. Rejecting unknown disturbances is what a control system does. No training is required. Perhaps somewhat less than impressive due to the choice of robotics platform, but:
Achieving natural behavior in a robot using neurally inspired hierarchical control Joseph W. Barter, Henry H. Yin
Paper: https://www.biorxiv.org/content/10.1101/2021.01.22.427862v1 https://www.biorxiv.org/content/10.1101/2021.01.22.427862v1
Movie 8: https://www.biorxiv.org/content/biorxiv/early/2021/01/25/2021.01.22.427862/DC8/embed/media-8.mp4?download=true https://www.biorxiv.org/content/biorxiv/early/2021/01/25/202...
Movie 9: https://www.biorxiv.org/content/biorxiv/early/2021/01/25/2021.01.22.427862/DC9/embed/media-9.mp4?download=true https://www.biorxiv.org/content/biorxiv/early/2021/01/25/202...
- jvanderbot 5y agoYeah, a lot of this tends to be "Let's try CNN on ... ", and it ends up doing about as well. It's bound to be a general purpose tool that's easy to use and doesn't require a classical approach. Like RRTs instead of visibility graphs for regular-old polygonal environments. The RRT is just easier vs hand-tuning obstacle boundaries, etc. I have some experience on this, as we are building a system like this with similar constraints, and there's really no "need" for anything other than classical control.