Learning Manifolds for Sequential Motion Planning

Motion planning with constraints is an important part of many real-world robotic systems. In this work, we study manifold learning methods to learn such constraints from data. We explore two methods for learning implicit constraint manifolds from data: Variational Autoencoders (VAE), and a new method, Equality Constraint Manifold Neural Network (ECoMaNN). With the aim of incorporating learned constraints into a sampling-based motion planning framework, we evaluate the approaches on their ability to learn representations of constraints from various datasets and on the quality of paths produced during planning.
Comment: Accepted for presentation at the Robotics: Science and Systems (RSS) 2020 Workshop for Learning (in) Task and Motion Planning. Paper length is 4 pages (i.e. 3 pages of technical content and 1 page of the references)