Model Predictive Path-Planning Controller With Potential Function for Emergency Collision Avoidance on Highway Driving

Abstract

Existing potential functions (PFs) utilized in autonomous vehicles mainly focus on solving the path-planning problems in some conventional driving scenarios; thus, their performance may not be satisfactory in the context of emergency obstacle avoidance. Therefore, we propose a novel model predictive path-planning controller (MPPC) combined with PFs to handle complex traffic scenarios (e.g., emergency avoidance when a sudden accident occurs). Specifically, to enhance the safety of the PFs, we developed an MPPC to handle an emergency case with a sigmoid-based safe passage embedded in the MPC constraints (SPMPC) with a specific triggering analysis algorithm on monitoring traffic emergencies. The presented PF-SPMPC algorithm was compiled in a comparative simulation study using MATLAB/Simulink and CarSim. The algorithm outperformed the latest PF-MPC approach to eliminate the severe tire oscillations and guarantee autonomous driving safety when handling the traffic emergency avoidance scenario.

Adaptive Potential Field with Collision Avoidance for Connected Autonomous Vehicles

“…We conducted a comprehensive co-simulation study using MATLAB/Simulink and CarSim simulator to verify the proposed ESPP-based APF method. To compare the effectiveness of the proposed method, we computed four types of path planners as follows: (i) conventional PF with constant speed (denoted as CPF with CS) [22]; (ii) adaptive PF with full braking maneuver (denoted as APF with FB) [19]; (iii) adaptive PF without the lower road PF when sensing the emergency (denoted as APF without LR); and (iv) ESPP-based adaptive PF (denoted as ESPP-based APF). The overall simulation was conducted on a laboratory laptop with Intel(R) Core(TM) i9-10980HK CPU@2.40GHz and RAM 32GB.…”

“…However, the road PF was set to be inviolable, and the speed was also assumed constant. Lin et al [22], [23] presented a unique safe tunnel-based model predictive pathplanning controller (STMPC) with the APF to solve the local minima problem that appeared on the highway; however, the longitudinal speed is assumed to be constant, and the road PF cannot be adjusted. To tackle the overtaking problem, Xie et al [24] presented a distributed motion planning framework via artificial PF to introduce the notion of velocity difference PF and acceleration difference PF for vehicle platoons.…”

“…However, considering the dynamic characteristics of the ego vehicle, it is difficult to stop at the local minima region owing to the large inertia. Instead, the ego vehicle could have driven into either the obstacle PF or road PF and received an excessive repulsive force [22], which can cause severe yawing and even vehicle crashes. A simple way is to directly remove the road PF (the side to which the vehicle is heading) so that the ego vehicle can drive out of the road to obtain sufficient space for emergency braking and obstacle avoidance.…”

An Emergency Escape Strategy based on Reasonable Dynamic Constraints for Autonomous Vehicles

“…Active obstacle avoidance is a basic requirement for an active escape algorithm, and various algorithms are developed to address this problem. Lin et al [1][2] mentioned *This work was supported by the National Key R&D Program of China, No: 2022YFB2503000…”

Energy Efficient Local Path Planning Algorithm Based on Predictive Artificial Potential Field

“…However, using the APF algorithm may result in the vehicle getting trapped at a local minimum [25]. Furthermore, measurement noise and imprecision in physical implementation may result in oscillations and generation of a non-smooth path [26].…”