MPC-Inspired Neural Network Policies for Sequential Decision Making
This work addresses the problem of scalable and efficient training of complex planning policies in continuous spaces for robotics or control applications, but it appears incremental as it builds on existing methods like DAgger and MPC.
The paper tackles training neural network policies for sequential decision making by extending the DAgger algorithm to handle MPC-inspired architectures, resulting in improved training performance, generalization, and robustness to disturbances and modeling errors.
In this paper we investigate the use of MPC-inspired neural network policies for sequential decision making. We introduce an extension to the DAgger algorithm for training such policies and show how they have improved training performance and generalization capabilities. We take advantage of this extension to show scalable and efficient training of complex planning policy architectures in continuous state and action spaces. We provide an extensive comparison of neural network policies by considering feed forward policies, recurrent policies, and recurrent policies with planning structure inspired by the Path Integral control framework. Our results suggest that MPC-type recurrent policies have better robustness to disturbances and modeling error.