Requirements and Test Cycles for Brake Systems of Autonomous Vehicle Concepts on the Example of an Autonomous Shuttle

Abstract

uring the development toward autonomous and electrified vehicles with low emissions, many visions for future mobility concepts arise, one of them being autonomous shuttles for urban areas. Most publications concerning these concepts focus on control and software while in this paper the change of requirements for wheel brakes is examined. The performance of wheel brakes for todays passenger cars is currently tested under different worst-case assumptions regarding area of operation and highest possible load resulting from human operation. Considering the capabilities of autonomous shuttles like autonomous driving and the availability of regenerative braking, these assumptions need to be reevaluated. This also includes comfort and lifetime requirements regarding wheel brakes for these concepts and takes in perspective that for an autonomous shuttle a certain area of operation is defined in their operational design domain (ODD) as well as a lower maximum velocity. To do so, different autonomous shuttle concepts are aggregated as well as their respective hardware and tech specs. To gather system requirements for the braking system of an autonomous shuttle a stakeholder analysis is performed, highlighting the underlying business model, driving tasks and passenger types as well as their needs and wishes. The shift in requirements is derived in comparison to conventional wheel brakes for cars. Usual performance tests for conventional wheel brakes for passenger cars are semantically analyzed to discuss their relevance and transferred into new performance tests for the given vehicle class. Three test scenarios are created, the first one being the “Emergency Braking Test”, which consists of two consecutive emergency brakings. Secondly a “Standard Operation Test” which consists of ten consecutive, comfortable accelerations and decelerations for passenger pickup and transport. Lastly, a “Hill Descent Test” on a long descent in the area of operation of the shuttle, like in the demanding urban topology of San Francisco. Based on the scenarios different availability levels of regenerative braking power are considered. Based on the developed test cycles a comparison is drawn for power and energy dissipation demand and the corresponding torques needed for an example vehicle under various levels of available regenerative braking power. While power and energy dissipation have decreased heavily, the torque demand is still as high as needed for a conventional vehicle. The changed requirements open up new possibilities for suitable braking concepts for autonomous shuttles. This may also reduce brake emissions depending on the chosen concepts.