The chassis frame of the self-propelled peanut seeder in hilly and mountainous areas is the main supporting structure of the entire machine, and its weight directly affects the operational performance of the seeder. Therefore, in response to the issues of structural heaviness, strength redundancy, and short endurance of the self-propelled peanut seeder in hilly and mountainous areas, this study aims to reduce the overall weight of the machine, conserve resources, and extend the seeder’s endurance time. The research focuses on the chassis frame of the self-propelled peanut seeder, utilizing SolidWorks for 3D modeling. A finite element model of the chassis frame is established using ANSYS Workbench, followed by modal analysis and static analysis under four different working conditions. Based on sensitivity analysis, design variables for the chassis frame are selected, and the response relationships between these design variables are simulated using the Latin Hypercube Design method combined with the Kriging approximation model. Finally, a multi-objective lightweight design is conducted based on the MOGA algorithm. The results indicate that the optimized chassis frame mass is reduced by 28.9%, while meeting the strength requirements. Field tests indicate that the plant spacing qualification rate is ≥98%; the seeding depth operational performance is stable, with an average qualification rate of seeding depth ≥90%. After lightweight design, the prototype structure is stable and the performance is reliable. The research results can provide reference and theoretical basis for the structural optimization and design of the walking chassis frame of self-propelled peanut planters in hilly and mountainous areas.

Aiming at the problem that the consistency and stability of seeding depth are difficult to control during the complex field operation of peanut seeders, this study proposes a control method combining the Kalman Filter algorithm based on Maximum Likelihood Estimation (MLE-KF) and the Fuzzy PID algorithm. A system was designed to automatically adjust the furrow opener during the operation of the peanut seeder, thereby realizing the automatic control of seeding depth. This method maintains the peanut seeding depth within a reasonable error range suitable for growth, and effectively ensures the consistency of peanut seeding depth. In this study, displacement sensors and ultrasonic sensors were used to directly or indirectly detect the furrow depth. Combined with electric push rods and actuators, the furrow depth was adjusted in real time during the field operation of the peanut seeder. For the filtering fusion algorithm, a noise parameter adaptive algorithm based on sliding window and approximate MLE was adopted, while a Fuzzy PID control strategy was employed for the controller. Simulations were conducted in MATLAB/Simulink, and the simulation results showed that the improved algorithm achieved a significant improvement in seeding depth accuracy compared with the traditional Kalman Filter and PID control. Field tests were carried out with the qualification rate as the index under different operating speeds and seeding depth conditions. For this seeding depth control method, when the operation speed was 3 km/h and the furrow depth was set to 50.00 mm, the qualification rate of seeding depth reached 94.80%, with a standard deviation of 3.12 mm and a coefficient of variation of 6.08%. Compared with the traditional mechanical adjustment and passive adjustment methods, this method increased the qualification rate by 4.60% and 3.30%, respectively, and reduced the coefficient of variation by 2.46% and 3.01%, respectively. The research on the seeding depth control method in this study provides a new detection and control approach for the seeding depth control of peanut seeders.