.png){kind=icon}
.png){kind=icon}
Return
Learning-Based Distributed MPC for Nonconvex Consensus Optimization With Collision Constraints
.png){kind=icon}
PRE
AI
DOI：10.1109/TASE.2025.3629680.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This article presents a novel approach to learning-based distributed model predictive control (LDMPC) for nonconvex optimization problems which aims to enhance the distributed system's consensus and avoid collision. Selecting the objective function of a distributed model predictive control (DMPC) system over a finite horizon to maximize performance and ensure safety is a challenging problem. The current work of this article is to introduce a function approximator that integrates DMPC and reinforcement learning (RL) through policy iteration (PI) to reconstruct the terminal cost function and reformulate the finite time nonconvex optimization problem. This work decouples the constraints and objective functions in the optimization process between multiple agents and introduces an improved alternating direction multiplier method (ADMM) as an consensus optimization solution of LDMPC. Moreover, the convergence, feasibility, and stability properties of our algorithm are proved in this article. The numerical example shows that the method updates can be performed distributively without inconsistency and demonstrates the effectiveness and safety of the LDMPC. Note to Practitioners-In practical engineering systems, the safety and stability of the agent are the keys to ensuring the safe and efficient operation of the system. Due to the complexity of the agent's dynamic and structural characteristics and the need for an accurate objective function, distributed model predictive control usually cannot determine the perfect objective function to simultaneously ensure safety and optimal performance. This article addresses the challenge of optimizing multi-agent systems in nonconvex environments, particularly focusing on constraints consensus and collision avoidance. This work introduces a LDMPC approach that integrates RL with DMPC to iteratively learn the terminal cost function and adopts an improved ADMM to solve the feasible solution, thereby improving system performance and safety. This approach is particularly useful in applications such as autonomous vehicles, robotics, and sensor networks, where collision avoidance and system consensus are critical. The method is data-driven, scalable, and does not require prior knowledge of the system's dynamics, making it adaptable to various scenarios. Future work could explore extending this framework to more complex systems with additional constraints and uncertainties.
Keywords:
Optimization
Linear programming
Convex functions
Convergence
Automation
Safety
Stability analysis
Predictive control
Multi-agent systems
Costs
Multi-agent system
distributed model predictive control
constrained consensus
reinforcement learning
alternating direction multiplier method
Journal
IF:
6.4
Papers:
4.9K
Citations:
1.6W