Autonomous mobile robots (AMRs) are becoming increasingly sophisticated, navigating complex environments and performing critical tasks. However, ensuring their safety and efficiency, especially during braking, remains a significant challenge. This paper introduces an innovative control framework that tackles these issues head-on, focusing on robust motion control, active fault tolerance, and efficient energy recovery.
TL;DR
This paper presents a novel nonlinear predictive motion control framework for autonomous mobile robots that integrates active fault-tolerant control and regenerative braking. The system enhances driving stability and braking safety, particularly during actuator failures, while also maximizing the recovery of braking energy.
What Problem Does This Paper Address?
Autonomous mobile robots (AMRs) rely on multiple actuators for steering, driving, and braking. This over-actuation, while offering flexibility, also introduces complexities in coordinated control and raises concerns about system reliability. A critical challenge is ensuring safe and efficient braking, especially when actuators fail. Traditional approaches often struggle to handle simultaneous actuator failures and regenerative braking, leading to reduced safety and missed opportunities for energy recovery. This research aims to bridge this gap by developing a control system that can adapt to actuator failures, maintain stability, and efficiently recapture braking energy.
Methods
The proposed framework employs a multi-layered control strategy:
- Longitudinal Motion Control: An integrated feedforward and feedback control algorithm is used for velocity tracking, designed to handle system nonlinearities and external disturbances.
- Lateral Motion Control (Path Tracking): Nonlinear Model Predictive Control (NMPC) is utilized for the four-wheel steering (4WS) path-tracking controller. NMPC allows for predicting future states and optimizing control actions over a horizon, leading to more precise path following.
- Active Fault-Tolerant Control (AFTC): This is a key innovation. When braking actuators (specifically in-wheel motors, IWMs) fail, the AFTC algorithm reallocates braking torques among the remaining functional actuators (both IWMs and electro-mechanical brakes, EMBs). This ensures continued braking capability and safety. The Weighted Least Squares (WLS) method is employed for this torque reallocation.
- Regenerative Braking: The system prioritizes using IWMs for braking to recover kinetic energy. EMBs are used to supplement the braking force when needed, especially in hybrid braking modes or during failures.
Datasets
Not discussed in the paper. The evaluation relies on simulations using the Carsim and Simulink co-simulation platform.
Key Insights and Ideas
- Hybrid Braking Synergy: Combining regenerative braking (IWMs) with mechanical braking (EMBs) offers the best of both worlds: improved braking safety and efficiency, shorter braking distances, and reduced braking times compared to purely regenerative or mechanical braking.
- AFTC for Robustness: The AFTC system is crucial for maintaining control during actuator failures. By intelligently redistributing braking torque, it prevents the AMR from deviating from its path and ensures safe stopping, even with multiple IWM failures.
- Prioritizing IWMs for Energy Recovery: The strategy of prioritizing IWMs for regenerative braking effectively maximizes energy recovery, contributing to the overall efficiency of the AMR.
- NMPC for Precise Path Following: The use of NMPC for path tracking allows the AMR to accurately follow its intended trajectory, even under dynamic conditions.
Key Results
The simulation results demonstrate the effectiveness of the proposed framework across three scenarios:
- Case 1 (No Failure): Hybrid braking significantly reduced braking distance and time compared to regenerative or mechanical braking alone, while still recovering substantial braking energy.
- Case 2 (Single/Double IWM Failure): The AFTC algorithm successfully compensated for the loss of one or two IWMs, allowing the AMR to maintain its path and brake safely. Without AFTC, the robot lost stability and deviated significantly from its path.
- Case 3 (Multiple IWM Failures): Even with the failure of three IWMs, the AFTC system enabled the AMR to continue braking safely and recover some braking energy, showcasing its robustness. The lateral offset without AFTC was notably larger than in Case 2.
Limitations and Future Work
The authors acknowledge that the current work primarily focuses on braking failures of IWMs. Future research directions include:
- Investigating hybrid conditions of IWM and EMB braking failures.
- Exploring the combined failure of IWM and EMB braking systems.
The study also relies on simulation; real-world validation would be a crucial next step.
Summary updated