Unsupervised Path Regression Networks
Michal Pándy, Daniel Lenton, Ronald Clark · 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · 2021
We demonstrate that challenging shortest path problems can be solved via direct spline regression from a neural network, trained in an unsupervised manner (i.e. without requiring ground truth optimal paths for training). To achieve this, we derive a geometry-dependent optimal cost function whose minima guarantees collision-free solutions. Our method beats state-of-the-art supervised learning baselines for shortest path planning, with a much more scalable training pipeline, and a significant speedup in inference time.