Model Predictive Control for Formation Placement and Recovery of Traffic Cone Robots

Abstract

The challenge of effectively managing the formation and recovery of traffic cone robots (TCRs) is addressed by proposing a linear time-varying model predictive control (MPC) strategy. This problem involves coordinating multiple TCR formations within a work area to reach a target location, which is a huge challenge due to the complexity of dynamic coordination. Unlike conventional approaches, our method decomposes the formation control problem into two main components: leader TCR motion planning and follower formation tracking control. The motion planning component involves path and velocity planning to achieve leader trajectory control, which serves as a reference trajectory for the follower. The formation tracking task extends to formation control among multiple robots to achieve the traffic cone robot formation placement and recovery task. To address the TCR input limitation problem, input constraints are considered during the design process of the MPC controllers. The effectiveness and practicality of the proposed control strategy are validated through a series of numerical simulations and physical experiments with TCRs.

Keywords

• traffic cone robots
• model predictive control
• formation control
• input constraints