Motorcycle riders are subject to a high risk of suffering severe or fatal injuries. Previous research has identified autonomous emergency braking for motorcycles (MAEB) as one of the most promising technologies to increase safety for riders (e.g., [2]).

Compared to drivers of two-track vehicles, emergency braking maneuvers are much more challenging for motorcyclists. As there is no restraint system such as a safety belt, riders need to support their upper body movement and they need to control and stabilize their vehicle. This requires attention, situation awareness and body tension. Before applying maximum deceleration, the rider has to achieve this ‘prepared-for-braking’ state. To generate optimal crash mitigation or even crash avoidance, the velocity should be reduced even before this state is achieved. Therefore, it is necessary to determine applicable preparatory braking profiles. As sudden unexpected braking maneuvers are critical for unprepared riders, there is still a great uncertainty on how high these decelerations can be. The identification of the limits would enable to determine the safety benefit of MAEB, when the full deceleration potential before reaching the ‘prepared-for-braking’ state is used. One of the main challenges in MAEB studies is the rider state. On one hand, to evaluate to what extent autonomous interventions can support riders, participants need to be unprepared to receive unbiased results. On the other hand, due to safety and ethical reasons, it is out of question to determine the limits of controllable decelerations with unprepared riders. For this purpose, the experiments within this project are split up:​

In a first study with experts, the deceleration limits are identified. The experts are asked to evaluate if different automatically applied braking interventions are controllable for unprepared average riders. By increasing the decelerations until the experts rate them as intolerable for unprepared riders, maximum tolerable decelerations for different braking profiles in real riding scenarios are defined.

In a following participant study, average riders experience a realistic emergency braking scenario (suddenly braking vehicle ahead). The deceleration profiles defined during the expert study are applied. With these experiments, the reaction of the unprepared participants to unexpected autonomous braking maneuvers are analyzed. The result is an evaluation on how partial braking maneuvers can help to reduce the transition time and on the potential decrease of velocity during the transition period.

In a third study, more critical scenarios (different secondary tasks) and the influence of warnings prior to the autonomous braking intervention are investigated on a dynamic motorcycle simulator.

The studies provide empirically obtained data on maximum deceleration values for different automatic braking interventions that are tolerable for average riders in unexpected emergency braking situations. The results also show the maximum amount of velocity – and thus kinetic energy – that can be reduced during the partial automatic braking phase before the maximum deceleration can be applied. The simulator experiments show the influence of different secondary tasks and the effect of visual-auditory warnings. The described method can be used as a reference for future development and configuration of MAEB.