This paper presents the design and field evaluation of a low-cost automatic steering system for agricultural tractors. The system employs a PID controller whose gains are tuned using a metaheuristic optimization method. Core hardware includes an ESP32 microcontroller, an MPU9250 inertial measurement unit, a GPS module, and a servo motor for closed-loop yaw angle control, with a complementary filter fusing gyroscope and magnetometer data for robust heading estimation. Nine optimization algorithms were systematically compared: Grid Search, Random Search, Bayesian Optimization, Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), Moth-Flame Optimization (MFO), Sine Cosine Algorithm (SCA), Whale Optimization Algorithm (WOA), and Salp Swarm Algorithm (SSA). A cost function combining overshoot and settling time was used. Step response analysis showed that WOA achieved the best performance, with an integral absolute error of 6. 31°·s, a settling time of 2. 15 s, and a minimal overshoot of 0. 08°. In field tests on asphalt and farmland, the WOA-tuned system reduced lateral deviation by 69% (from 12. 4 cm to 3. 8 cm) and 67% (from 18. 7 cm to 6. 2 cm), respectively, compared to manual steering. Repeated-measures ANOVA and paired t-tests confirmed statistically significant improvements (p 2. 7). The core components cost under 150 USD. The study offers a reproducible pipeline for comparative metaheuristic evaluation in agricultural vehicle guidance.
Karamolachab et al. (2026) studied this question.