An Augmentation Function for Active Pedestrian Safety System Based on Crash Risk Evaluation

This study proposed an augmentation function to current active pedestrian safety systems (APSSs), which is expected to be effective for pedestrian crashes caused by pedestrians’ unexpected behavior. The augmentation function estimates the crash risk with a pedestrian given its time-space-distance relationship with the pedestrian; the crash risk represents the probability of hitting that pedestrian given all the pedestrian's possible random trajectories in the near future. Once the crash risk exceeds a toleration threshold, the augmentation function activates evasive actions, even if there is no current conflict with the pedestrian. A Monte Carlo process was used to estimate the crash risk under different sets of vehicle and pedestrian kinematic features. The possible pedestrian trajectories were sampled from a fine-tuned Markovian integrated random walk model; in particular, kinematic variations between pedestrian types were considered. Then, the effectiveness of evasive actions was evaluated. It was found that children and young/middle-aged pedestrians require higher-intensity evasive actions in zones of moderate and severe crash risk; while old pedestrians need higher-intensity evasive actions in zones of mild crash risk. In addition, it is necessary to select proper speed reduction rate and combination of lane change and speed, to reduce the crash risk. Finally, the study demonstrated that a field of view of 50° and a detection range of 40 m would be the minimum requirement to support the augmentation function, which requires an upgrade for many automobile manufacturers’ current APSSs.

[1]  Aniket S. Ahire.  Night Vision System in BMW , 2014 .

[2]  Thomas A. Dingus,et al.  Forward-Looking Collision Warning System Performance Guidelines , 1997 .

[3]  Thierry Serre,et al.  Accident simulation and reconstruction for enhancing pedestrian safety: issues and challenges , 2014 .

[4]  Andreas Eidehall,et al.  Benefit estimation model for pedestrian auto brake functionality , 2010 .

[5]  Trevor Darrell,et al.  PANDA: Pose Aligned Networks for Deep Attribute Modeling , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Leopoldo Armesto,et al.  An Active Safety System for Low-Speed Bus Braking Assistance , 2017, IEEE Transactions on Intelligent Transportation Systems.

[7]  Sunan Huang,et al.  Evaluation of remote pedestrian sensor system based on the analysis of car-pedestrian accident scenarios , 2008 .

[8]  Xiaoou Tang,et al.  Pedestrian Attribute Recognition At Far Distance , 2014, ACM Multimedia.

[9]  Van-Dung Hoang,et al.  Pedestrian Action Prediction Based on Deep Features Extraction of Human Posture and Traffic Scene , 2018, ACIIDS.

[10]  Saeid Nahavandi,et al.  Early intent prediction of vulnerable road users from visual attributes using multi-task learning network , 2017, 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[11]  Leopoldo Armesto,et al.  Haptic Feedback to Assist Bus Drivers for Pedestrian Safety at Low Speed , 2016, IEEE Transactions on Haptics.

[12]  Kaiqi Huang,et al.  Multi-attribute learning for pedestrian attribute recognition in surveillance scenarios , 2015, 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR).

[13]  Wassim G Najm,et al.  Crash Avoidance Needs and Countermeasure Profiles for Safety Applications Based on Light-Vehicle-to-Pedestrian Communications , 2016 .

[14]  Antonella Ferrara,et al.  Onboard Sensor-Based Collision Risk Assessment to Improve Pedestrians' Safety , 2007, IEEE Transactions on Vehicular Technology.

[15]  Azim Eskandarian,et al.  Handbook of Intelligent Vehicles , 2012 .

[16]  Serge P. Hoogendoorn,et al.  Simulation of pedestrian flows by optimal control and differential games , 2003 .

[17]  Robert Anderson,et al.  Description of Pedestrian Crashes in Accordance with Characteristics of Active Safety Systems , 2014 .

[18]  Nils Lubbe,et al.  Brake reactions of distracted drivers to pedestrian Forward Collision Warning systems. , 2017, Journal of safety research.

[19]  Koji Suzuki,et al.  Development of Pre-Crash Safety System with Pedestrian Collision Avoidance Assist , 2013 .

[20]  Patrick Heinemann,et al.  Context-based detection of pedestrian crossing intention for autonomous driving in urban environments , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[21]  Kip Smith,et al.  Pedestrian injury mitigation by autonomous braking. , 2010, Accident; analysis and prevention.

[22]  Nerijus Kudarauskas.  Analysis of Emergency Braking of a Vehicle , 2007 .

[23]  Robert Anderson,et al.  Issues and challenges for pedestrian active safety systems based on real world accidents. , 2015, Accident; analysis and prevention.

[24]  Michael P. Clamann,et al.  Automated Vehicles and Pedestrian Safety: Exploring the Promise and Limits of Pedestrian Detection. , 2019, American journal of preventive medicine.

[25]  Dariu Gavrila,et al.  Active Pedestrian Safety by Automatic Braking and Evasive Steering , 2011, IEEE Transactions on Intelligent Transportation Systems.

[26]  Christophe F. Wakim,et al.  A Markovian model of pedestrian behavior , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[27]  Mohamed Abdel-Aty,et al.  In-depth approach for identifying crash causation patterns and its implications for pedestrian crash prevention. , 2020, Journal of safety research.

[28]  Klaus C. J. Dietmayer,et al.  Early detection of the Pedestrian's intention to cross the street , 2012, 2012 15th International IEEE Conference on Intelligent Transportation Systems.