Learning to Generate Cost-to-Go Functions for Efficient Motion Planning
This addresses efficiency issues in motion planning for practical robots, offering a novel approach that could enhance real-time robotic applications.
The paper tackles the computational burden of traditional motion planning by introducing a neural network architecture that directly generates cost-to-go functions, enabling faster trajectory generation without extensive iterative propagation or collision checking. Simulation results show c2g-HOF is orders of magnitude faster at execution time than methods exploring configuration space during execution.
Traditional motion planning is computationally burdensome for practical robots, involving extensive collision checking and considerable iterative propagation of cost values. We present a novel neural network architecture which can directly generate the cost-to-go (c2g) function for a given configuration space and a goal configuration. The output of the network is a continuous function whose gradient in configuration space can be directly used to generate trajectories in motion planning without the need for protracted iterations or extensive collision checking. This higher order function (i.e. a function generating another function) representation lies at the core of our motion planning architecture, c2g-HOF, which can take a workspace as input, and generate the cost-to-go function over the configuration space map (C-map). Simulation results for 2D and 3D environments show that c2g-HOF can be orders of magnitude faster at execution time than methods which explore the configuration space during execution. We also present an implementation of c2g-HOF which generates trajectories for robot manipulators directly from an overhead image of the workspace.