Mats Jonasson

Control of electric vehicles with autonomous corner modules : implementation aspects and fault handling

In this paper, vehicle dynamics for electric vehicles equipped with in-wheel motors and individual steering actuators are studied adopting the principles of optimal tyre-force allocation. A simple method for describing the constraints owing to tyre and actuator limitations is described. The control architecture is evaluated by investigating its response to realistic fault conditions. The evaluation demonstrates that the control architectures ability to ensure vehicle stability generally is good. However, during major faults and extreme driving situations, vehicle stability is not maintained unless the constraints in the optimisation process used for tyre-force allocation are adapted to the specific fault.

Closed-loop Controller for Post Impact Vehicle Dynamics Using Individual Wheel Braking and Front Axle Steering

This paper presents a vehicle path controller for reducing the maximum lateral deviation (Ymax) after an initial impact in a traffic accident. In previous research, a Quasi-Linear Optimal Controller (QLOC) was proposed and applied to a simple vehicle model with individually controlled brake actuators. QLOC uses non-linear optimal control theory to provide a semiexplicit approximation for optimal post-impact path control, and in principle can be applied to an arbitrary number of actuators. The current work extends and further validates the control method by analysing the effects of adding an active front axle steering actuator at different post-impact kinematics, as well as increasing the fidelity of the vehicle model in the closed-loop controlled system. The controller performance is compared with the results from openloop numerical optimisation which uses the same vehicle model. The inherent robustness properties of the QLOC algorithm are demonstrated by its direct application to an independent high-fidelity multi-body vehicle model. Towards real-time implementation, the algorithm is further simplified so that the computational efficiency is enhanced, whereas the performance is shown not to be degraded.

Utilisation of optimisation solutions to control active suspension for decreased braking distance

This work deals with how to utilise active suspension on individual vehicle wheels in order to improve the vehicle performance during straight-line braking. Through numerical optimisation, solutions have been found as regards how active suspension should be controlled and coordinated with friction brakes to shorten the braking distance. The results show that, for the studied vehicle, the braking distance can be shortened by more than 1u2005m when braking from 100u2005km/h. The applicability of these results is studied by investigating the approach for different vehicle speeds and actuator stroke limitations. It is shown that substantial improvements in the braking distance can also be found for lower velocities, and that the actuator strokes are an important parameter. To investigate the potential of implementing these findings in a real vehicle, a validated detailed vehicle model equipped with active struts is analysed. Simplified control laws, appropriate for on-board implementation and based on knowledge of the optimised solution, are proposed and evaluated. The results show that substantial improvements of the braking ability, and thus safety, can be made using this simplified approach. Particle model simulations have been made to explain the underlying physical mechanisms and limitations of the approach. These results provide valuable guidance on how active suspension can be used to achieve significant improvements in vehicle performance with reasonable complexity and energy consumption.

Using future path information for improving the stability of an over-actuated vehicle

In this paper, model predictive control (MPC) is applied for controlling an over-actuated vehicle. The control problem is associated with the distribution of the tyre forces to ensure vehicle stability. The use of MPC is shown to be a suitable method if the vehicles future desired trajectory is known. Simulation studies conducted show that access to information in advance, even if such information is restricted to only a few seconds, significantly contributes to maintaining vehicle stability. Furthermore, a longer prediction horizon results in earlier actions and stabilises the vehicle even better.

Control allocation strategies for an electric vehicle with a wheel hub motor failure

Three fault-tolerant control strategies for electric vehicles with wheel hub motors are presented and compared, which are all based on the control allocation principle. The main objective is to maintain the directional stability of the vehicle in case of a component failure during high speed manoeuvres. Two simplified strategies that are suited for on-board implementation are derived and compared to an optimal control allocation strategy and a reference vehicle with a basic electronic stability control system. The occurring faults are considered to be in the electric high-voltage system that can arise in wheel hub motors. All three control allocation strategies show improved re-allocation of traction forces after a severe fault, and hence an improved directional stability. However, the performance of both simplified algorithms shows limitations in case of force demands outside the capabilities of the respective actuator. This work shows that vehicle safety is increased by the proposed fault-tolerant control strategies.

Road friction effect on the optimal vehicle control strategy in two critical manoeuvres

This paper presents a research study on the optimal way to negotiate safety-critical vehicle manoeuvres depending on the available actuators and road friction level. The motive is to provide viable knowledge of the limitations of vehicle capability under the presence of environmental preview sensors. In this paper, an optimal path is found by optimising the sequence of actuator requests during the manoeuvres. Particular attention is paid to how the vehicle control strategy depends on friction. This study shows that the actuation of all the forces and torques on and around the vehicle centre of gravity is approximately scaled with friction, whereas at individual wheel level, the optimal force allocation will differ under different friction conditions. A lower friction level leads to lower velocities and load transfer, which influences the individual wheels’ tyre force constraints. However, the actuator response compared to the whole system is increased at a lower friction level.

Steering redundancy for self-driving vehicles using differential braking

ABSTRACT This paper describes how differential braking can be used to turn a vehicle in the context of providing fail-operational control for self-driving vehicles. Two vehicle models are developed with differential input. The models are used to explain the bounds of curvature that differential braking provides and they are then validated with measurements in a test vehicle. Particular focus is paid on wheel suspension effects that significantly influence the obtained curvature. The vehicle behaviour and its limitations due to wheel suspension effects are, owing to the vehicle models, defined and explained. Finally, a model-based controller is developed to control the vehicle curvature during a fault by differential braking. The controller is designed to compensate for wheel angle disturbance that is likely to occur during the control event.