Combining High Level Scheduling and Low Level Control to Manage Fleets of Mobile Robots

TL;DR

The paper addresses scalable coordination of fleets of mobile robots in industrial settings by coupling a high-level scheduling layer (ComSat) with a low-level distributed MPC controller. ComSat generates time-parameterized routes that are bridged to execution by a local planner and a distributed MPC that handles obstacle and inter-robot collision avoidance. Experimental results in simulated 2D environments of varying scale demonstrate high task completion rates, tight adherence to schedules, and resilience to congestion and disruptions, with a safety margin of $\mu=20$ s to prevent conflicts. The framework is modular and extensible, offering a practical path toward real-world deployment and potential enhancements in energy optimization and human-robot interaction.

Abstract

The deployment of mobile robots for material handling in industrial environments requires scalable coordination of large fleets in dynamic settings. This paper presents a two-layer framework that combines high-level scheduling with low-level control. Tasks are assigned and scheduled using the compositional algorithm ComSat, which generates time-parameterized routes for each robot. These schedules are then used by a distributed Model Predictive Control (MPC) system in real time to compute local reference trajectories, accounting for static and dynamic obstacles. The approach ensures safe, collision-free operation, and supports rapid rescheduling in response to disruptions such as robot failures or environmental changes. We evaluate the method in simulated 2D environments with varying road capacities and traffic conditions, demonstrating high task completion rates and robust behavior even under congestion. The modular structure of the framework allows for computational tractability and flexibility, making it suitable for deployment in complex, real-world industrial scenarios.

Figures (6)

• Figure 1: Visualization of an illustrative environment, some roads are wide enough for bidirectional traffic or overtaking in the same direction, which is not possible for other narrow roads. The plant can be abstracted into a graph, as shown in the right sub-figure.
• Figure 2: Flowchart of ComSat roselli2024conflict.
• Figure 3: Map of the small environment with four horizontal and four vertical narrow road segments. The map is abstracted into a graph (nodes from N01 to N34 and orange edges), and the robots are represented by circles, with their reference trajectories marked by crosses ($\times$) of different colors and predicted trajectories marked by pluses ($+$).
• Figure 4: Execution delays (in Seconds) of the four robots in the small environment. The delay, positive when the robot is late or negative otherwise, is the difference between the actual arrival time at a node versus the scheduled arrival time.
• Figure 5: Delay distribution for ten robots in the large environment.