Driver Intention Algorithm for Pedestrian Protection and Automated Emergency Braking Systems

摘要: Driver intention recognition reveals an enormous potential for automated driving. Current emergency braking (AEB) systems warn and take over vehicle control in case of frontal collision risks. It is high interest to detect driver's before initiating such transitions driving order avoid annoyance. Also the number warnings can be reduced when driver indicates a safe behavior planning. Especially pedestrian protection will benefit urban environments where pedestrians potentially enter trajectory frequently. The paper describes development detection algorithm systems. data collected simulator study with 32 participants, balanced roughly by gender, age style. Every participant circulated environment random traffic. In six scenarios were was either on road or walked towards road. An system would activate warning eventually situations. This could adapted cancelled existing driver. A fixed base Face Lab Eyetracking device stored gaze patterns relation pedestrians' position presentation screen. Pedal activity also recorded. analysis both sources allowed rule based brake due pedestrian. proposes sequential eye pedal presented discussed this paper. correlated highly reference method developed earlier Diederichs Pohler [1]. 94 % match their results.