Liu Jingxin, Liao Xiaofeng, Dong Jin-Song, Mansoori Amin
College of Electronic and Information Engineering, Southwest University, Chongqing 400715, China; School of Computing, National University of Singapore, Singapore 117417, Singapore.
Key Laboratory of Dependable Services Computing in Cyber-Physical Society (Chongqing) Ministry of Education, College of Computer, Chongqing University, Chongqing 400044, China.
Neural Netw. 2023 Apr;161:693-707. doi: 10.1016/j.neunet.2023.02.011. Epub 2023 Feb 15.
This paper investigates a class of power consumption minimization and equalization for intelligent and connected vehicles cooperative system. Accordingly, a distributed optimization problem model related to power consumption and data rate of intelligent and connected vehicles is presented, where the power consumption cost function of each intelligent and connected vehicle may be nonsmooth, and the corresponding control variable is subject to the constraints generated by data acquisition, compression coding, transmission and reception. We propose a distributed subgradient-based neurodynamic approach with projection operator to achieve the optimal power consumption of intelligent and connected vehicles. By differential inclusion and nonsmooth analysis, it is confirmed that the state solution of neurodynamic system converges to the optimal solution of the distributed optimization problem. With the help of the algorithm, all intelligent and connected vehicles asymptotically reach a consensus on an optimal power consumption. Simulation results show that the proposed neurodynamic approach is capable of effectively solving the problem of power consumption optimal control for intelligent and connected vehicles cooperative system.
本文研究了一类智能网联车辆协作系统的功耗最小化与均衡问题。相应地，提出了一个与智能网联车辆的功耗和数据速率相关的分布式优化问题模型，其中每辆智能网联车辆的功耗成本函数可能是非光滑的，且相应的控制变量受到数据采集、压缩编码、传输和接收所产生的约束。我们提出一种基于分布式次梯度的带有投影算子的神经动力学方法，以实现智能网联车辆的最优功耗。通过微分包含和非光滑分析，证实了神经动力学系统的状态解收敛到分布式优化问题的最优解。借助该算法，所有智能网联车辆在最优功耗上渐近达成共识。仿真结果表明，所提出的神经动力学方法能够有效解决智能网联车辆协作系统的功耗最优控制问题。
