Automated Braking Decision and Control for Pedestrian Collision Avoidance Based on Risk Assessment
Authors:
- 17 August 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Intelligent Transportation Systems Magazine
- Vol. 14 (3) , 41-61
Abstract
An automated braking decision and control framework is proposed in this article to avoid intelligent vehicle–pedestrian collisions. The framework contains three parts: pedestrian collision risk assessment, automated braking decision making, and automated braking control. First, a kinematic vehicle model and a Kalman filter are used to estimate the path trajectories of a host vehicle and a pedestrian, respectively, by using onboard sensor information. Then, pedestrians with the highest risk levels are selected as the most dangerous collision objects from the valid target pedestrian sets. Second, based on road adhesion conditions, vehicle lateral stability conditions, and driver intention constraints, an automated braking decision strategy is proposed using fuzzy theory; in this approach, the collision time difference is the input, and the desired deceleration is the output. Third, a first-order, inertial pure hysteresis braking system response model is built based on real vehicle braking experimental data. The relationship between braking deceleration and the master cylinder pressure is established through data fitting to achieve automated braking control for the host vehicle. Finally, experiment results indicate that the proposed control framework can calculate reasonable deceleration rates and perform braking control to achieve complete collision avoidance, thus verifying the effectiveness of the algorithm.Keywords
Funding Information
- National Natural Science Foundation of China (62003238, 62073152, U19A2069)
This publication has 30 references indexed in Scilit:
- Probabilistic risk assessment for pedestrian–vehicle collision considering uncertainties of pedestrian mobilityTraffic Injury Prevention, 2017
- Estimation of driving style in naturalistic highway traffic using maneuver transition probabilitiesTransportation Research Part C: Emerging Technologies, 2017
- Design of Pedestrian target selection with Funnel Map for Pedestrian AEB SystemIEEE Transactions on Vehicular Technology, 2016
- Driver and pedestrian awareness-based collision risk analysis2016
- Estimation of walking direction for pedestrian path prediction from moving vehicle2015
- A Controlled Interactive Multiple Model Filter for Combined Pedestrian Intention Recognition and Path Prediction2015
- A Novel Kalman Filter with State Constraint Approach for the Integration of Multiple Pedestrian Navigation SystemsMicromachines, 2015
- Robust multiple cameras pedestrian detection with multi-view Bayesian networkPattern Recognition, 2015
- Bayesian Human Motion Intentionality Prediction in urban environmentsPattern Recognition Letters, 2014
- Pedestrian fatality risk as a function of car impact speedAccident Analysis & Prevention, 2009