Learning-based Model Predictive Control for Safe Exploration and Reinforcement Learning

Abstract

Reinforcement learning has been successfully used to solve difficult tasks in complex unknown environments. However, these methods typically do not provide any safety guarantees during the learning process. This is particularly problematic, since reinforcement learning agent actively explore their environment. This prevents their use in safety-critical, real-world applications. In this paper, we present a learning-based model predictive control scheme that provides high-probability safety guarantees throughout the learning process. Based on a reliable statistical model, we construct provably accurate confidence intervals on predicted trajectories. Unlike previous approaches, we allow for input-dependent uncertainties. Based on these reliable predictions, we guarantee that trajectories satisfy safety constraints. Moreover, we use a terminal set constraint to recursively guarantee the existence of safe control actions at every iteration. We evaluate the resulting algorithm to safely explore the dynamics of an inverted pendulum and to solve a reinforcement learning task on a cart-pole system with safety constraints.

14 pages, 7 figures. arXiv admin note: substantial text overlap with arXiv:1803.08287

References in corpus (5)

• Safe Model-based Reinforcement Learning with Stability Guarantees
• Safe Exploration in Continuous Action Spaces
• Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control
• Lyapunov-based Safe Policy Optimization for Continuous Control
• Constrained Policy Optimization

Cited by in corpus (18)

• Safe Reinforcement Learning Using Robust MPC
• Data-driven Economic NMPC using Reinforcement Learning
• Stochastic data-driven model predictive control using Gaussian processes
• Online learning-based Model Predictive Control with Gaussian Process Models and Stability Guarantees
• Robust Learning-based Predictive Control for Discrete-time Nonlinear Systems with Unknown Dynamics and State Constraints
• End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks
• Wasserstein Robust Reinforcement Learning
• Lagrangian-based online safe reinforcement learning for state-constrained systems
• Posterior Variance Analysis of Gaussian Processes with Application to Average Learning Curves
• Safe Interactive Model-Based Learning
• Hybrid Gaussian Process Modeling Applied to Economic Stochastic Model Predictive Control of Batch Processes
• Safety Guarantees for Planning Based on Iterative Gaussian Processes
• Safety Reinforced Model Predictive Control (SRMPC): Improving MPC with Reinforcement Learning for Motion Planning in Autonomous Driving
• Safely Learning Dynamical Systems from Short Trajectories
• ABC-LMPC: Safe Sample-Based Learning MPC for Stochastic Nonlinear Dynamical Systems with Adjustable Boundary Conditions
• Nested Mixture of Experts: Cooperative and Competitive Learning of Hybrid Dynamical System
• Distributed Safe Learning using an Invariance-based Safety Framework
• Safe Wasserstein Constrained Deep Q-Learning