Related Experiment Video
Updated: May 8, 2026
07:28
A Contusive Model of Unilateral Cervical Spinal Cord Injury Using the Infinite Horizon Impactor
Published on: July 24, 2012
19.6K
Trajectory planning framework for autonomous vehicles based on collision injury prediction for vulnerable road users
Yage Guo1, Yu Liu2, Botao Wang1
1State Key Laboratory of Advanced Design and Manufacturing Technology for Vehicle, Hunan University, Changsha 410082, China.
Accident; Analysis and Prevention
|May 15, 2024
Summary
This study models Electric Two-Wheeler (E2W) collisions to predict head injuries (HIC15) and develops an autonomous vehicle trajectory planning framework prioritizing vulnerable road user (VRU) safety. The goal is to minimize E2W rider head injuries in potential autonomous vehicle interactions.
Area of Science:
- Road safety engineering
- Autonomous driving systems
- Biomechanics and injury analysis
Background:
- Accidents involving Electric Two-Wheelers (E2Ws) are a growing safety concern, exacerbated by the rise of autonomous driving technology.
- Existing trajectory planning for autonomous vehicles often neglects the safety of vulnerable road users (VRUs), such as E2W riders.
- Understanding the kinematic response of E2Ws in collisions is crucial for developing effective safety measures.
Purpose of the Study:
- To investigate the kinematic response of E2Ws in vehicle collisions and analyze the impact of collision parameters on head injuries (HIC15).
- To establish robust injury prediction models for E2W riders in various collision scenarios.
- To propose an innovative trajectory planning framework for autonomous vehicles that prioritizes VRU head injury prediction and risk mitigation.
Main Methods:
- Developed a multi-rigid-body model of E2W-vehicle collisions, validated against real accident data.
- Created a large-scale crash dataset using a parameterized simulation framework and Monte Carlo sampling.
- Employed MLP + XGBoost regression for developing the head injury prediction model and validated the trajectory planning algorithm in simulated scenarios.
Main Results:
- The validated multi-rigid-body model demonstrated high accuracy (≤11% error) in reconstructing accident scenarios.
- The MLP + XGBoost injury prediction model achieved a high R² of 0.92 on the test set, indicating strong predictive performance.
- The proposed trajectory planning framework effectively integrated VRU head injury prediction for enhanced autonomous driving safety.
Conclusions:
- The study successfully established reliable models for E2W collision kinematics and head injury prediction.
- The developed trajectory planning framework offers a novel approach to enhance autonomous vehicle safety for VRUs.
- This research contributes to safer integration of autonomous vehicles by prioritizing the protection of vulnerable road users like E2W riders.
