Brake architecture considerations for maximising regenerative braking In contemporary mobility, the trend of electrification has led to significant increases in braking system architectural complexity. While it is clear that overall vehicle efficiency has regained focus in the minds of the product developers and consumers, brake energy recovery remains the most significant efficiency technology. Many different approaches to braking interfaces are now offered in electrified vehicles, and while the overall technological implementations have common aspects, often the depth of regenerative braking can vary between vehicles. Research Objectives In this paper, a review of contemporary electric vehicles will be presented, showing the depth and performance of regenerative braking offered. Where applicable, novel HMI elements are also considered (e.g. accelerators, paddles, settings). The paper will then present a review of the relevant brake architectures for these vehicles, and highlight both the operational aspects as well as the failure tolerant design approaches. Further to this, the paper will suggest how significant fault injections can be handled, and how conceptual/architectural choices are important in determining performance and safety limits. Methodology A simple market review was used to select relevant benchmark vehicles. Test program content was selected to objectively compare regenerative braking performance in the context of overall braking performance. Overall braking performance was established in reference to ECE R13H standard, with adaptions to examine the role of regenerative braking. Test topics considered included objective and subjective pedal feel, static tests, effectiveness, high speed fade, coast down, stopping distance, parking brake, brake failures. For individual vehicles, additional tests were performed to characterise any novel HMI elements. Comparisons of the architecture for each vehicle involved research methods, physical component checks, and systematic analysis of the performance observed during the test program. The brake failure tests in particular were useful to establish redundancy approaches within the observed architectures. In considering fault tolerance for regenerative brake systems, a number of critical scenarios were developed with the aim of highlighting the vehicle dynamics effects on fault occurrence in regenerative braking. Results The results obtained from the benchmarking tests were gathered and analysed using well-established (semi-automated) post processing. A comprehensive report for each tested vehicle was created, and using these reports, the team were able to select specific test results of interest to further examine. While many similarities were found in terms of vehicle response to brake pedal inputs, significant differences in off-pedal braking were observed. OEMs must conform to stringent regulatory performance requirements for (brake pedal) braking, but a diversity of approaches is evident in terms of HMI concepts which result in vehicle deceleration. Limitations The authors had significant access to contemporary vehicles and suitable test facilities. However, brake system level interactions (such as network messages) could not be directly observed, only their vehicle level outcomes. As such, the handling and resolution of critical situations within the brake system relies on careful consideration of vehicle performance, the author’s prior knowledge, and relevant literature. Further, vehicle testing was non-destructive, whereby each vehicle needed to be returned to its original condition at the end of the test schedule. Actuation hardware was not disassembled nor individually tested, but instead literature research was used to establish internal architecture. What’s New? The paper will present some brake architecture layout criteria for handling primary failure scenarios, including additional aspects relevant for regenerative systems. These criteria will then be linked to depth of regenerative braking, in order to clearly explain the link between fault tolerant brake architecture and overall vehicle efficiency. Conclusions The work described by this paper shows a number of common regenerative brake architectural choices employed across a variety of global OEMs. The braking performance of the considered vehicles is easily comparable under objective and formal test, but significant subjective differences are present for both braking and HMI aspects. The performance requirements of such systems directly relates to both the conventional and novel braking controls, both during normal and fault-tolerant operation.

Mr. Deaglán Ó Meachair, Braking Consultant, BrakeBetter Ltd; Mr. Fabio Squadrani, Brakes Manager, Idiada; Mr. Narcis Molina, Brakes Engineer, Idiada