Path Planning with Neural Networks
페이지 정보
작성자 최고관리자 댓글 조회 작성일 21-11-03 17:14본문
- Tamar, A., Wu, Y., Thomas, G., Levine, S., and Abbeel, P. Value iteration networks. In NIPS, 2016.
- Lee, L., Parisotto, E., Chaplot, D. S., Xing, E., and Salakhutdinov, R. Gated path planning networks. In ICML, 2018.
- Chaplot, Devendra Singh, Deepak Pathak, and Jitendra Malik. "Differentiable Spatial Planning using Transformers." International Conference on Machine Learning. ICML, 2021.
We consider the problem of spatial path planning. In contrast to the classical solutions which optimize a new plan from scratch and assume access to the full map with ground truth obstacle locations, we learn a planner from the data in a differentiable manner that allows us to leverage statistical regularities from past data. We propose Spatial Planning Transformers (SPT), which given an obstacle map learns to generate actions by planning over long-range spatial dependencies, unlike prior data-driven planners that propagate information locally via convolutional structure in an iterative manner. In the setting where the ground truth map is not known to the agent, we leverage pre-trained SPTs in an end-to-end framework that has the structure of mapper and planner built into it which allows seamless generalization to out-of-distribution maps and goals. SPTs outperform prior state-of-the-art differentiable planners across all the setups for both manipulation and navigation tasks, leading to an absolute improvement of 7-19%.
첨부파일
- 211028_pp_with_nn.pdf (2.6M) 27회 다운로드 | DATE : 2021-11-03 17:14:24
댓글목록
등록된 댓글이 없습니다.