Research on fault prediction and speed control system for unmanned combine harvesters based on IPSO-SVM and fuzzy logic.

This study proposes an IPSO-SVM-based fault prediction and fuzzy speed control system for unmanned combine harvesters. The primary goal is to prevent clogging failures and ensure long-term stable operation of unmanned harvesting machines, maintaining efficiency while minimizing downtime. The system integrates multi-component slip rate data, collected from critical parts of the harvester, and uses the IPSO-SVM model for fault warning. The fuzzy control algorithm adjusts the operating speed based on the predicted fault status and feeding rate to mitigate clogging risks. Experimental results show that the system can accurately identify over 98.5% of fault states and reduce the occurrence of complete blockage by adjusting the harvester's speed within 0.5 to 2 seconds after minor clogging. This work demonstrates the feasibility of applying the system in field environments, providing a reliable solution for the intelligent and unmanned operation of combine harvesters in fields.