Parameter Optimization of Model Predictive Direct Motion Control for Distributed Drive Electric Vehicles Considering...

Area of Science:

• Automotive Engineering
• Control Systems
• Artificial Intelligence

Background:

• Distributed drive electric vehicles (DDEVs) require advanced control strategies for simultaneous longitudinal and lateral motion management.
• Integrating vehicle dynamics, motor control, and driver comfort presents a significant challenge in electric vehicle (EV) control systems.

Purpose of the Study:

• To develop and validate a novel model predictive direct motion control (MPDMC) strategy for DDEVs.
• To enhance vehicle performance by simultaneously optimizing longitudinal/lateral motion, energy efficiency, and driving experience.
• To investigate the effectiveness of particle swarm optimization (PSO) and long short-term memory (LSTM) neural networks in refining control parameters.

Main Methods:

• Analysis of the DDEV dynamic model, incorporating vehicle body and in-wheel motor characteristics.
• Development of an MPDMC approach using a single CPU to generate voltage references based on a minimized cost function.
• Construction of a cost function considering longitudinal velocity, yaw rate, lateral displacement, efficiency, and driving feel, optimized via PSO and LSTM-based driving feel evaluation.

Main Results:

• The optimized MPDMC strategy demonstrated superior performance compared to baseline methods in simulations.
• The integrated approach effectively balanced vehicle dynamics, energy efficiency, and subjective driving quality across "Normal", "Eco", and "Sport" modes.
• MATLAB and CarSim simulations validated the MPDMC strategy's effectiveness in four distinct driving scenarios.

Conclusions:

• The proposed MPDMC strategy offers a robust and efficient solution for controlling DDEVs.
• The combination of MPC, PSO, and LSTM provides a powerful framework for optimizing complex vehicle control objectives.
• This research contributes to the advancement of intelligent vehicle control systems, enhancing both performance and driver satisfaction.