Model Predictive Path Integral Control using Covariance Variable Importance Sampling

TL;DR

This paper introduces a GPU-accelerated Model Predictive Path Integral control algorithm that employs a generalized importance sampling scheme, allowing for adaptive drift and diffusion modifications to improve control performance.

Contribution

It presents a novel importance sampling approach for MPPI control that adapts drift and diffusion, enhancing parallel optimization efficiency on GPUs.

Findings

Enhanced control performance with generalized importance sampling

Successful GPU-based parallel optimization implementation

Comparison shows advantages over differential dynamic programming

Abstract

In this paper we develop a Model Predictive Path Integral (MPPI) control algorithm based on a generalized importance sampling scheme and perform parallel optimization via sampling using a Graphics Processing Unit (GPU). The proposed generalized importance sampling scheme allows for changes in the drift and diffusion terms of stochastic diffusion processes and plays a significant role in the performance of the model predictive control algorithm. We compare the proposed algorithm in simulation with a model predictive control version of differential dynamic programming.