Model Predictive Path Integral Control using Covariance Variable Importance Sampling
This work addresses control optimization for robotics or autonomous systems, but it appears incremental as it builds on existing MPPI methods with a modified sampling approach.
The paper tackles the problem of improving Model Predictive Path Integral (MPPI) control by introducing a generalized importance sampling scheme that modifies drift and diffusion terms, and implements it on a GPU for parallel optimization. The result is a performance comparison in simulation with a model predictive control version of differential dynamic programming, though no concrete numbers are provided.
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.