A Data-Driven Method Towards Minimizing Collision Severity for Highly Automated Vehicles

Abstract

The deployment of autonomous vehicles on public roads calls for the development of methods that are reliably able to mitigate injury severity in case of unavoidable collisions. This study proposes a data-driven motion planning method capable of minimizing injury severity for vehicle occupants in unavoidable collisions. The method is based on establishing a metric that models the relationship between impact location and injury severity using real accident data, and subsequently including it in the cost function of a motion planning framework. The vehicle dynamics and associated constraints are considered through a precomputed trajectory library, which is generated by solving an optimal control problem. This allows for efficient computation as well as an accurate representation of the vehicle. The proposed motion planning approach is evaluated by simulation, and it is shown that the trajectory associated with the minimum cost mitigates the collision severity for occupants of passenger vehicles involved in the collision.

Vehicle Crash Mitigation Strategy in Unavoidable Collision Scenarios: Focusing on Motion Planning by Considering a Generalized Crash Severity Model

“…[10] handles the collision mitigation planning using Crash Severity Maps from multiple offline FEM simulations, which contain different combinations of lateral offset and relative heading angles between vehicles. To minimize collision severity in planning, a data-driven model is presented via a trajectory library relating injury severity with impact location, which comes from real accident data [11]. The effectiveness of such learning-based models depends on how well the crash data can cover the real application scenarios, especially when the data come from real accident surveys.…”

“…Then the final cost is expressed as the squared error between the desired and the observed outputs. Finally, the tracking problem of the CSI-optimal path is formulated and solved as a linear MPC problem by combining ( 9) to (11).…”

Vehicle Motion Planning for Crash Mitigation in Unavoidable Collision Scenarios

“…On the other hand, most motion planning algorithms focus on how to avoid obstacles in potentially dangerous scenarios [4][5][6][7][8][9][10][11], while very few have contributed to how to reduce the crash injury in unavoidable collisions via motion planning [12][13][14][15][16]. For the example in Fig.…”

Human injury-based safety decision of automated vehicles

“…Inherent highly nonlinear characteristics lead to a lack of accurate mathematical methods to interpret an injury process explicitly. A few simplified indicators have been used to represent crash severity (e.g., impact location or vehicle body deformation), which cannot characterize human injuries well due to the oversimplification of human-vehicle system interactions ( Wang et al., 2019 ; Simon et al., 2019 ; Parseh et al., 2021 ). To solve the need for injury prediction, one promising solution is to combine existing traffic crash information as a databank with proper mining tools ( Delen et al., 2017 ; Li et al., 2019 ; Wang et al., 2021 ).…”