Distributed Non-Convex Model Predictive Control for Non-Cooperative Collision Avoidance of Networked Differential Drive Mobile Robots

Abstract

This paper proposes a novel Non-Cooperative Distributed Model Predictive Control (NC-DMPC) strategy for collision avoidance of networked multiple Differential Drive Mobile Robots (multi-DDMRs). This strategy is based on Priority-Based NC-DMPC (PB-NC-DMPC). To avoid the loss of the prediction consistency caused by PB-NC-DMPC, a concept of urgency is introduced to DDMRs in the Networked Control Systems (NCS), and a corresponding priority queue is obtained by sorting these urgency values. Renumbering these priorities in descending order converts the partial order priorities into a topological order, which is a valid sequence for the DDMRs to solve their optimization problems. At last, the Sequential Convex Programming (SCP) is presented to solve the non-convex optimization problem resulting from the collision avoidance constraints. As a special application, the proposed strategy solves the problem of collision avoidance of Multi-DDMRs, compared to the original PB-NC-DMPC, the proposed strategy offers a more effective and safe performance in complex applications.

Keywords

• Collision avoidance
• differential drive mobile robots
• networked control system
• model predictive control