Self-Triggered MPC for Teleoperation of Networked Mobile Robotic System via High-Order Estimation

Abstract

Since the teleoperation process of networked mobile robotic systems (NMRSs) is generally affected by the coupling of the human force, complex dynamical model, external environment and nonholonomic constraints, designing an effectiveness control to guarantee the desired performance for the teleoperation task is a challenging work. Besides, due to the limitation of the communication and control technology, the controller for the teleoperation system is usually required to have low computation load and low monitoring sampling frequency. To address these challenges, a novel self-triggered model predictive control (STMPC) framework, being consisted of the local STMPC, high-order estimation and M/R fixed-time controller, is constructed. Compared to traditional MPC methods, our proposed local STMPC offers lower computational load and requires fewer monitoring resources while retaining the ability to optimize control performance and handle multiple constraints. Using the presented STMPC framework in a hierarchical manner and newly designed high-order estimation, we successfully and simultaneously solve the teleoperation task while accounting for the self-triggering mechanism, decentralized control implementation, disturbance rejection, and complex nonholonomic model. Additionally, we derive sufficient conditions for ensuring the stability of the closed-loop system. Finally, the simulation and experiment results are provided to demonstrate the effectiveness of the proposed STMPC framework. Note to Practitioners—Our research introduces a groundbreaking framework, self-triggered model predictive control (STMPC), specifically designed to elevate control efficiency, quality, and reliability in the context of remote operation of networked mobile robotic systems (NMRSs). This innovation holds immense promise, particularly in domains requiring the utilization of autonomous vehicle fleets, such as large-scale exploration, search and rescue missions, and escort operations in hazardous or hard-to-reach areas (e.g., forest firefighting, warzone patrolling, saturation attacks, and space escort). These environments are often characterized by extreme external conditions and unpredictable external influences, such as meteorite impacts, flames, obstacles, and explosions, posing significant challenges for the control and task execution of unmanned vehicle fleets. Moreover, these missions demand high precision and control responsiveness from the robotic clusters to swiftly accomplish urgent tasks. Consequently, the need arises for autonomous vehicle fleets that not only possess the capacity to adapt to various constraints and external disturbances but also execute mission tasks swiftly and precisely. STMPC, as presented in this paper, demonstrates its capability to address these issues and requirements effectively, making it a versatile and powerful tool for enhancing control and system performance in various scenarios.