Model Predictive Control-Based Value Estimation for Efficient Reinforcement Learning

doi:10.1109/MIS.2024.3386204

Model Predictive Control-Based Value Estimation for Efficient Reinforcement Learning

2024-05-01

0

OA

AI

DOI：10.1109/MIS.2024.3386204

Original

Original request for help

Save

Abstract

En 中文

Reinforcement learning (RL) suffers from limitations in real practices primarily due to the number of required interactions with virtual environments. It results in a challenging problem because we are implausible to obtain a local optimal strategy with only a few attempts for many learning methods. Hereby, we design an improved RL method based on model predictive control that models the environment through a data-driven approach. Based on the learned environment model, it performs multistep prediction to estimate the value function and optimize the policy. The method demonstrates higher learning efficiency, faster convergent speed of strategies tending to the local optimal value, and less sample capacity space required by experience replay buffers. Experimental results, both in classic databases and in a dynamic obstacle-avoidance scenario for an unmanned aerial vehicle, validate the proposed approaches.

Keywords:

Predictive models

Neural networks

Trajectory

Data models

Computational modeling

Training

Optimization

Reinforcement learning

Predictive control

Model Predictive Control-Based Value Estimation for Efficient Reinforcement Learning

Abstract

Journal

Organization