FISITA World Congress 2023 – Abstract Research on the Evaluation of Pedal Misapplication Prevention Device Based on Driver’s Pedal Work Ryotaro Kai*, Shoko Oikawa, Toshiya Hirose, Shibaura Institute of Technology, Japan KEYWORDS – ADAS, Driving simulator, Human machine interface, Pedal misapplication, Acceleration Suppression Objective There is a method that aims to reduce traffic accidents by detecting a pedal misapplication and inhibiting vehicle acceleration when a driver controls the pedal by mistake. The purpose of this study was to evaluate the operation of a Pedal Misapplication Prevention Device (PMPD) that used pedal stroke speed as an activation condition, and to determine the optimal threshold value. Methodology A driving simulator used in this study was equipped with a cylindrical screen, a 6-axis motion device, and a turntable. We used the pedal stroke speed as the operating condition for PMPD. By setting the threshold of PMPD activation based on the measurement results of the driver’s pedal work, we verified the effect of PMPD on the suppression of acceleration against mis-step and the effect on the false activation. The pedal stroke speed value was monitored every 0.05 seconds, and acceleration inhibition was activated when the pedal stroke speed was above the threshold value. There were three main types of experiments. Experiment 1 evaluated whether the PMPD functioned properly on a linear map. Experiment 2 investigated whether unintended activation of the system occurred during normal driving in an urban area. Experiment 3 investigated PMPD activation on highways. For subject evaluation, a questionnaire survey was conducted. Results In Experiment 1, when starting and re-acceleration, the acceleration control was activated during abrupt pedal operation, of which occurrence rate was at 50%. When the pedal was operated at full stroke, the acceleration control was successfully activated about 96%. In the misstep test, acceleration suppression was activated about 81.3% when the pedal speed was not normal, such as pedaling quickly or at full stroke. On the other hand, in the misstep test at the same pedal operation speed as the normal brake pedal, the acceleration suppression was activated only 25%. In Experiment2 for unwanted activation in the city, there was no unwanted activation. In Experiment 3, the acceleration suppression was activated only about 45% of the time during abrupt pedal operation on the highway. Limitations of this study The experimental participants in this study were young adults only. Pedal missteps also tend to be seen among the elderly, so one of our future tasks is to conduct experiments for elderly participants. What does the paper offer that is new in the filed in comparison to the other works of the author? Currently, PMPDs for low speeds are in practical use, but it’s desirable to construct PMPDs for higher speeds in the future. In this study, the system operation is determined based on the driver’s pedal work, and its effectiveness is confirmed by simulator experiments even at high speed. The accelerator pedal may be operated as if it were the brake pedal. Therefore, this study investigates not only the speed of the accelerator pedal, but also the speed of the brake pedal. In addition, this study also investigates changes in acceleration suppression performance as affected by changes in pedal depressing speed thresholds. Conclusion Acceleration suppression was confirmed in the experiments that assumed missteps such as abrupt pedal operation and full pedal operation. While the system was able to detect abrupt missteps, there were still issues in detecting gradual missteps. Therefore, the threshold values are being reexamined to investigate the changes in acceleration suppression performance given by changes in the threshold values for operating conditions.

Ryotaro Kai, Student, Shibaura Institute of Technology