RL-RRT: Kinodynamic Motion Planning via Learning Reachability Estimators from RL Policies — mentioned by andreyk

Sun, 22 Aug 2021 10:02:56 GMT

I think this is a bit too cynical/negative. It's not true that neural net approaches don't work, they just have different strengths and weaknesses (strength being generalization/flexibility/ease of use, weakness being long term coherence, interpretability, reliability). There are plenty of examples of traditional methods being combined with neural net methods (eg https://ai.stanford.edu/blog/qagnn/ , https://arxiv.org/abs/1907.04799 , NeRF), and I think that is likely to happen in this context as well to some extent.

Paper and discussion history. A mention is not an endorsement.

RL-RRT: Kinodynamic Motion Planning via Learning Reachability Estimators from RL Policies · HackerPapers discussions

RL-RRT: Kinodynamic Motion Planning via Learning Reachability Estimators from RL Policies — mentioned by andreyk