Cost-to-Go Function Generating Networks for High Dimensional Motion Planning

TL;DR

This paper introduces c2g-HOF networks that learn to generate cost-to-go functions for high-dimensional manipulator motion planning, enabling faster and smoother trajectory generation directly from workspace sensor data.

Contribution

The paper proposes a novel neural network architecture combining cost-to-go and higher order functions, trained end-to-end for efficient motion planning in complex environments.

Findings

Planning with c2g-HOF is significantly faster than traditional algorithms.

c2g-HOF generalizes well to generate smoother, lower-cost trajectories.

Trajectory planning in complex workspaces takes only 0.13 seconds.

Abstract

This paper presents c2g-HOF networks which learn to generate cost-to-go functions for manipulator motion planning. The c2g-HOF architecture consists of a cost-to-go function over the configuration space represented as a neural network (c2g-network) as well as a Higher Order Function (HOF) network which outputs the weights of the c2g-network for a given input workspace. Both networks are trained end-to-end in a supervised fashion using costs computed from traditional motion planners. Once trained, c2g-HOF can generate a smooth and continuous cost-to-go function directly from workspace sensor inputs (represented as a point cloud in 3D or an image in 2D). At inference time, the weights of the c2g-network are computed very efficiently and near-optimal trajectories are generated by simply following the gradient of the cost-to-go function. We compare c2g-HOF with traditional planning algorithms for various robots and planning scenarios. The experimental results indicate that planning with c2g-HOF is significantly faster than other motion planning algorithms, resulting in orders of magnitude improvement when including collision checking. Furthermore, despite being trained from sparsely sampled trajectories in configuration space, c2g-HOF generalizes to generate smoother, and often lower cost, trajectories. We demonstrate cost-to-go based planning on a 7 DoF manipulator arm where motion planning in a complex workspace requires only 0.13 seconds for the entire trajectory.