Abstract: To achieve steering control for a complex wheeled ground robot, a fuzzy genetic PID control method was applied. First, a fuzzy neural network was used to establish the vehicle model, then genetic PID was used for parameter optimization, and finally, the optimized parameters were used to control the robot's steering. This method can intuitively determine whether the PID parameters are effective. Experimental studies on vehicle steering control showed good control performance. Keywords: Ground robot; Neural network; PID; Steering control 1 Overview The system parameters of a wheeled robot change significantly with variations in speed, steering angle, and other parameters, making accurate modeling difficult. Precise steering control of the vehicle is challenging. Based on the characteristics of wheeled robots, this paper proposes a method for establishing a fuzzy artificial neural network model. First, the speed is fuzzified, then a large multi-node artificial neural network is broken down into multiple sub-neural networks at different speeds. These sub-neural networks are connected using fuzzy control methods. Multiple consecutive given experiments can be performed on each sub-neural network, allowing for extensive training to ensure high accuracy for each sub-neural network. This method significantly reduces computational load and is suitable for practical applications. For details, please click: Experimental Study on Fuzzy Genetic PID Steering Control of Wheeled Robots