Abstract
Constrained robot motion planning is a widely used technique to solve complexrobot tasks. We consider the problem of learning representations of constraintsfrom demonstrations with a deep neural network, which we call EqualityConstraint Manifold Neural Network (ECoMaNN). The key idea is to learn alevel-set function of the constraint suitable for integration into aconstrained sampling-based motion planner. Learning proceeds by aligningsubspaces in the network with subspaces of the data. We combine both learnedconstraints and analytically described constraints into the planner and use aprojection-based strategy to find valid points. We evaluate ECoMaNN on itsrepresentation capabilities of constraint manifolds, the impact of itsindividual loss terms, and the motions produced when incorporated into aplanner.