TY - JOUR
T1 - Anticipation-Based Autonomous Platoon Control Strategy with Minimum Parameter Learning Adaptive Radial Basis Function Neural Network Sliding Mode Control
AU - Negash, Natnael M.
AU - Yang, James
N1 - Publisher Copyright:
© 2022 SAE International.
PY - 2022/4/25
Y1 - 2022/4/25
N2 - This article investigates the headway and optimal velocity tracking of autonomous vehicles (AVs), considering their predictive driving for the stability and integrity of spatial vehicle formation in the platoon. First, the human-like anticipation car-following model is used for modeling the autonomous system. Second, an adaptive radial basis function neural network (ARBF-NN)-based sliding mode control (SMC) is proposed for the control purpose. The control objective is to regulate traffic perturbation during entire road operations. To enable the controller to experience less computational burden and adaptation complexity, a minimum parameter learning (MPL) has also been integrated with ARBF-NN-based SMC. Third, an illustrative simulation example has been performed for two scenarios, i.e., constant headway and time-varying headway of vehicles. A performance comparison between the proposed controller and the conventional SMC was conducted, and controller parameter sensitivity was also carried out. The simulation results show that the proposed controller is an effective and ingenious method for platoon system control compared to the conventional sliding mode controller. Parameter sensitivity analysis shows that only three parameters need greater attention for maximum convergence rate and disturbance attenuation. The parameters c, Æž, and k can alter the responses of the vehicles.
AB - This article investigates the headway and optimal velocity tracking of autonomous vehicles (AVs), considering their predictive driving for the stability and integrity of spatial vehicle formation in the platoon. First, the human-like anticipation car-following model is used for modeling the autonomous system. Second, an adaptive radial basis function neural network (ARBF-NN)-based sliding mode control (SMC) is proposed for the control purpose. The control objective is to regulate traffic perturbation during entire road operations. To enable the controller to experience less computational burden and adaptation complexity, a minimum parameter learning (MPL) has also been integrated with ARBF-NN-based SMC. Third, an illustrative simulation example has been performed for two scenarios, i.e., constant headway and time-varying headway of vehicles. A performance comparison between the proposed controller and the conventional SMC was conducted, and controller parameter sensitivity was also carried out. The simulation results show that the proposed controller is an effective and ingenious method for platoon system control compared to the conventional sliding mode controller. Parameter sensitivity analysis shows that only three parameters need greater attention for maximum convergence rate and disturbance attenuation. The parameters c, Æž, and k can alter the responses of the vehicles.
KW - Autonomous platoon
KW - adaptive radial basis function
KW - anticipation-based control strategy, minimum parameter learning
KW - neural network
KW - sliding mode control
UR - http://www.scopus.com/inward/record.url?scp=85131220984&partnerID=8YFLogxK
U2 - 10.4271/10-06-03-0017
DO - 10.4271/10-06-03-0017
M3 - Article
AN - SCOPUS:85131220984
SN - 2380-2162
VL - 6
JO - SAE International Journal of Vehicle Dynamics, Stability, and NVH
JF - SAE International Journal of Vehicle Dynamics, Stability, and NVH
IS - 3
ER -