Search Results

The aims of this study are to investigate the responses of a Hybrid III dummy and a human body model in rollover crashes of an SUV, and to assess the effect of seat belts on occupant kinematics in rollover events. A SAEJ2114 rollover test of a 1994 Ford Explorer for two front row dummies with an inflatable tubular structure (ITS) is reconstructed and validated in MADYMO. By removing the ITS, the simulations of the Hybrid III dummy occupants with and without seat belts are obtained. By replacing the dummy models with human body models, with and without seat belts, two other combinations are also modelled. The kinematics and injury risks of two kinds of occupant models are compared and evaluated. Significant differences exist in the motions, and injury levels of the dummies and human body models with and without seat belts. Seat belts can significantly mitigate against occupant ejection.

Motivated by applying artificial intelligence technology to the automobile industry, reinforcement learning is becoming more and more popular in the community of intelligent driving research. The reward function is one of the critical factors which affecting reinforcement learning. Its design principle is highly dependent on the features of the agent. The agent studied in this paper can do perception, decision-making, and motion-control, which aims to be the assistant or substitute for human driving in the latest future. Therefore, this paper analyzes the characteristics of excellent human driving behavior based on the six-layer model of driving scenarios and constructs it into a human knowledge graph. Furthermore, for highway pilot driving, the expert demo data is created, and the reward function is self-learned via inverse reinforcement learning. The reward function design method proposed in this paper has been verified in the Unity ML-Agent environment.