Active Limit Handling for Enhanced Passenger Vehicle Safety

Lead Research Organisation: Cranfield University
Department Name: Sch of Engineering

Abstract

Road traffic injuries have become a leading cause of death globally accounting for 1.2 million deaths annually, and will rise in worldwide rank to sixth place as a major cause of death (including decease), by 2020. It is encouraging that, despite the constant increase of the number of vehicles in Europe during the last decade, the number of fatalities demonstrates a slow decay. This can be partly attributed to the enormous improvements in vehicle safety, through the introduction of both passive and active safety systems. By no means, however, have the current state-of-the-art of vehicle safety systems proven adequate to radically reverse the sober traffic accident statistics.

Current active safety systems, such as the Electronic Stability Control (ESC), aim at restricting the operation of the vehicle within a region characterised by an on-demand linear increase of tyre forces, away from the tyre's maximum force capacity, allowing the average driver to maintain control of the vehicle. With this project we wish to explore the benefits of using the whole of the available performance of the vehicle, rather than restricting its response, in accident avoidance situations. We propose the development of novel control algorithms, which will use the control authority introduced by current active safety systems and modern power/drive-train configurations, and employ expert driving skills to actively assist the driver exploit the limits of handling of the vehicle during emergency manoeuvring. MIRA, one of the world's leading independent providers of vehicle product engineering, testing, certification and research, has expressed their great interest in exploring the limits of the handling capacity of vehicles with modern power/drive-train configurations and the potential benefits in active safety. The company has agreed to offer their support to this project by means of technical consultation and active participation in the management and execution of the proposed research tasks.

Current drive-by-wire (DBW) actuators have allowed for a considerably enhanced control authority over the vehicle, as compared to traditional steering, brake and power/drive-train systems. The human operator provides commands through the conventional controls, that is, the steering wheel, and the throttle and brake pedals, whereas, for instance, the ESC allows for individual wheel braking, and electric motors in hybrid vehicles allow for individual wheel torque control. Race drivers have developed expert techniques to exploit most of the available force capacity of the tyres using the traditional controls. The enhanced control authority provided by modern vehicle controls potentially allows for even more efficient use of the available tyre performance. In this work we wish to explore the performance limits of modern vehicles equipped with DBW actuators, and identify optimum operating conditions related to accident avoidance. The first research task of the proposed work is to obtain steady-state cornering conditions at the limit of handling of the vehicle, that is, in a region of vehicle operation where the tyres produce forces close to their maximum capacity. As a case study we will consider the power/drive-train configuration of MIRA's prototype hybrid vehicle (H4V) which uses two high-torque independently controlled electric motors to drive the rear wheels. Consequently, we will design controllers using linear and nonlinear control design tools, which will stabilise the vehicle in potentially unstable driving conditions, instead of restricting the vehicle in a stable operating region away from its performance limits, using DBW control inputs. The control design will be implemented in a high fidelity simulation environment using experimentally validated vehicle models provided by MIRA. The implementation strategy entails the detection of emergency situations and accounts for the driver's intention.

Planned Impact

Road traffic injuries (RTI's) have become a leading cause of death globally accounting for 1.2 million deaths annually, and will rise in worldwide rank to sixth place as a major cause of death (including decease), by 2020. In 2002 RTI's resulted in 127,000 fatalities in Europe and an estimated 2,540,000 injuries, and accounted for the 16% of deaths from injuries. This type of injuries is currently the leading cause of death in Europe in the 5-29 years age group. Besides the social ramifications of these sober statistics, the directly measurable cost of road accidents in the EU countries is of the order of EUR 45 billion, while the indirect costs (including physical and psychological damage suffered by the victims and their families) are three to four times higher. The combined annual figure is equivalent to 2% of the EU's GNP. It is encouraging that, despite the constant increase of the number of vehicles - both freight and passenger - in Europe during the last decade, the number of fatalities demonstrate a slow decay. This can be partly attributed to the enormous improvements in vehicle safety, through the introduction of both passive and active safety systems. Passive safety features aim at minimising the consequences of an accident to the passengers of the vehicles or the pedestrians involved (seat belts, air-bags, controlled-deformation bodywork, etc). In this project we focus on active safety systems, which use sensors to estimate the vehicle state with respect to the environment and drive-by-wire actuators to provide control commands (brake, steering, throttle) and assist the driver to maintain control of the vehicle, or intervene in cases of driver loss of concentration or limited situational awareness. By no means, however, have the current state-of-the-art of passive and active safety systems proven adequate to radically reverse the above statistics.

Considering the number of deaths and injuries resulting from road traffic accidents and the associated financial implications the proposed research has the potential for major socio-economic impact. Pursuing this innovative research programme can radically reduce car accident rates to levels much lower than it is possible by current active safety systems. In order for the public to benefit from these technological advancements, the commercialisation of the research outcomes of the project is necessary, which would also contribute to further economic growth of automotive manufactures and OEMs, and attract research and development investment from global businesses. The proposed collaboration between Brunel University and MIRA will have an immediate benefit on the research capacity and knowledge of both parties. Finally, we envision that the public can benefit from this research even before the transfer of the new technologies to commercial passenger vehicles, by raising the awareness and understanding of the major causes of traffic accidents, as well as the limitations of current active and passive safety systems.

Publications

10 25 50

Related Projects

Project Reference Relationship Related To Start End Award Value
EP/I037792/1 23/04/2012 18/11/2012 £99,821
EP/I037792/2 Transfer EP/I037792/1 18/05/2013 17/04/2014 £56,622
 
Description The aim of the project is to extend newly acquired methodologies to characterize the vehicle behavior near the limit of handling (operating near the maximum available tyre forces) in the design of active chassis systems, taking advantage of the enhanced control authority available through modern drive-by-wire actuators. As a case study we consider a prototype hybrid vehicle developed by MIRA which offers rear axle torque vectoring capabilities for stability control and propulsion.

At this stage of the project we are developing a control architecture which uses the actuation layout of MIRA's vehicle to provide simultaneous cornering stability and velocity regulation during cornering near the maximum available lateral acceleration. The controller essentially tracks a reference trajectory which is generated according to the driver's steering input. The velocity regulation function ensures the feasibility of the reference trajectory, while stability is maintained with respect to near limit operating conditions. The control architecture will be validated in simulations and compared against traditional yaw stability control approaches.
Exploitation Route The proposed control architecture hinges on existing and emerging actuation technologies in passenger vehicles; see for instance the drivetrain configuration of hybrid/electric vehicles. The implementation of the control algorithms involves solely software development and hence the associated cost will be minimal. Therefore the proposed research and expected outcomes can be realistically put into production by vehicle manufacturers and OEM's.
In order for the public to benefit from these technological advancements, the commercialization of the research outcomes of the project is necessary, which would also contribute to further economic growth of automotive manufactures and OEMs, and attract research and development investment from global businesses.
Sectors Transport

 
Description Collaboration with MIRA 
Organisation Mother and Infant Research Activities (MIRA)
Country Nepal 
Sector Charity/Non Profit 
PI Contribution We provided insight on the use of advance optimisation and control theory in the development of active chassis systems.
Collaborator Contribution MIRA is the industrial partner to this project providing validated simulation models of a test vehicle for validation of the proposed control architecture. MIRA also provides guidance throughout the project through regular meetings of the project steering committee.
Impact The following paper was a result from our interaction with MIRA: E. Siampis, M. Massaro and E. Velenis, "Electric Rear Axle Torque Vectoring for Combined Yaw Stability and Velocity Control Near the Limit of Handling", IEEE Conference on Decision and Control, Florence, Italy, December 10-13, 2013.
Start Year 2012
 
Description Collaboration with TU Delft 
Organisation Delft University of Technology (TU Delft)
Country Netherlands 
Sector Academic/University 
PI Contribution PhD student co-supervision
Collaborator Contribution PhD student from Delft working on joint publications.
Impact The following publication resulted from this collaboration: D. Katzourakis, E. Velenis, E. Holweg and R. Happee, "Haptic Steering Support for Driving Near the Vehicle's Handling Limits: Test-Track Case", IEEE Transactions on Intelligent Transportation Systems, Vol. 15, No. 4, pp. 1781-1789, 2014.
Start Year 2009
 
Description Collaboration with University of Padova 
Organisation University of Padova
Country Italy 
Sector Academic/University 
PI Contribution Exchange student co-supervision.
Collaborator Contribution A PhD student from the University of Padova visited for 6 months Brunel University and Cranfield University and worked on several publications.
Impact The following papers resulted form this collaboration: D. Tavernini, M. Massaro, E. Velenis, D. Katzourakis and R. Lot, "Minimum Time Cornering: The Effect of Road Surface and Car Transmission Layout", Vehicle System Dynamics, Vol. 51, No. 10, pp. 1533-1547, 2013 D. Tavernini, E. Velenis, R. Lot and M. Massaro, "On the Optimality of Handbrake Cornering", IEEE Conference on Decision and Control, Florence, Italy, December 10-13, 2013.
Start Year 2012
 
Title Torque Vectoring controller for combined longitudinal/lateral vehicle dynamics stabilisation 
Description A control algorithm using rear axle electric torque vectoring to track a reference yaw rate generated by the driver's steering command, and simultaneously regulate vehicle's speed in cases where the requested yaw rate exceeds the physical limits of the vehicle. 
Type Of Technology Software 
Year Produced 2012 
Impact Use of multivariable control to address combined longitudinal lateral vehicle dynamics control near the limit of handling, as reported in relevant publication. 
 
Title Vehicle Dynamics model capturing the behavior of the vehicle near the limit of handling 
Description Matlab script which solves equations of motion of a vehicle in planar motion. It includes nonlinear tyre model, coupling of tyre forces in longitudinal and lateral directions and weight transfer effects. 
Type Of Technology Software 
Year Produced 2010 
Impact The vehicle model is able to capture vehicle behavior during the application of expert driving techniques, e.g. rally driving, as reported in relevant publications.