Lyapunov-based fault tolerant control allocation

Abstract

This work addresses the design of an active fault-tolerant motion controller for overactuated road vehicles. Our concept revolves around a control allocation framework, extended with Lyapunov-based stability constraints and costs. Besides the ability to cope with actuator constraints and optimal actuator reconfiguration, the proposed Lyapunov-based control allocation (LCA) allows the graceful degradation of control performance. Theoretical analysis and simulation results demonstrate that, in comparison with classical control allocation, the proposed LCA reduces tracking errors of the motion controller in the aftermath of actuator faults. Experimental tests carried out with the ROboMObil, a robotic electric vehicle prototype with wheel-based steering and traction actuation, demonstrate the effectiveness of the LCA

Toward Fault-Tolerant Vehicle Motion Control for Over-Actuated Automated Vehicles: A Non-Linear Model Predictive Approach

“…Multiple publications present fault-tolerant vehicle motion control approaches for over-actuated topologies featuring wheel-individual all-wheel steering and wheel-individual allwheel drive and/or brake actuators [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35]. However, apart from our previous work [3], no MPC-based approach is known to us that directly controls four steering and four drive/brake actuators with a single integrated control layer for fault-tolerant vehicle motion control.…”

“…The upper control layer usually calculates generalized forces in the vehicle's center of gravity, which are distributed to the wheels by means of different control allocation techniques at the lower control layer. For the upper control layer, researchers use sliding mode control [8], [9], [10], [11], [12], classic control techniques, such as PI [13], [14], [15], [16], [17] or PD combined with feedforward control [18], [19], [20], flatness-based control [21], H∞ control [22], robust state feedback control enhanced by feed-forward control [23], [24], LQR control [7], [25], as well as MPC [26], [27], [28], [29]. Control allocation techniques at the lower control layer mostly solve an optimal control problem, either online [8], [9], [10], [12], [14], [15], [22], [23], [24], [26], [27], [28], [29], [30] or analytically [7], [11], [16], [18], [19], [20], [21],…”

“…With respect to brake and drive actuators, zero wheel torque (F1), which is addressed in [7], [8], [9], [11], [12], [14], [15], [18], [19], [20], [21], [23], [24], [25], [26], [27], [28], [29], [31], [32], [35], and [36] is most commonly considered. Other degradations and failures of brake and drive actuators are reduced torque ranges (D1) [10], [15], [16], [17], [23], [24], [34], [35], [37], locked wheels (F3) [18], [19], [20], [21], [25], and unintended torques (F2) [13], [14], [37]. For steering actuators, an unintended steering angle (F4) is most commonly considered [7], [8], [14], [17], [18], [19], [20], [21],…”

“…Other degradations and failures of brake and drive actuators are reduced torque ranges (D1) [10], [15], [16], [17], [23], [24], [34], [35], [37], locked wheels (F3) [18], [19], [20], [21], [25], and unintended torques (F2) [13], [14], [37]. For steering actuators, an unintended steering angle (F4) is most commonly considered [7], [8], [14], [17], [18], [19], [20], [21], [23], [24], [25], [33]. Additionally, a reduced steering angle range (D2) [15], [16], [23], [24], [34], [38], zero steering torque (F5) [7], [15], [25], [31], [32], [39], and reduced steering dynamics (D3) [15] are found in the literature.…”

IEEE VTS Motor Vehicles Challenge 2023: A Multi-physical Benchmark Problem for Next Generation Energy Management Algorithms

“…This last feature is particularly attractive to improve tracking performance and vehicle dynamics stability of path/trajectory following algorithms of autonomous vehicles [12]. It also offers redundant traction actuators that can be exploited by fault tolerant controllers to improve vehicle reliability [13].…”