Advancements in Wheelie Phenomena Control Strategies for Enhanced Motorcycle Performance and Safety

This Ph.D. thesis addresses the challenge of the development of an innovative Anti- Wheelie (AW) system for high-performance commercial vehicles, where the occurrence of a front wheel lift can induce severe safety and performance implications. Current stateof- the-art AW systems often suffer from detection delays, necessitating the development of innovative solutions to enhance vehicle stability and control. This research focuses on the conception and validation of two novel AW systems tailored to speciĄc vehicle applications. The Ąrst AW system is designed for production vehicles, operating under the constraints of standard sensor suites, limited computational resources, and adaptability to diverse external conditions. This system innovation lies in a wheelie-preventing action through feedback and a model-based feedforward controller, aimed at stabilizing pitch dynamics before a wheelie event occurs. It employs a triggering component that considers both longitudinal and lateral dynamic quantities, together with explicit interaction with rider torque request. Additionally, a wheelie angle controller, integrated through state-of-theart detection techniques, enhances system safety. Conversely, the second AW system utilizes Non-Linear Model Predictive Control (NMPC) and is tailored for track-like travel, incorporating specialized hardware to meet increased computational demands. The NMPC-based system assumes knowledge of road slope, is based on a wheelie-related internal model, and includes explicit modeling of actuation delays, critical in preventing wheelie events. It is implemented on embedded hardware using Raspberry-Pi modules and is designed for commercial vehicles speciĄcally designed for tracks. Both controllers undergo rigorous testing in simulated and experimental scenarios. Simulations are conducted for initial parameter tuning and compatibility assessment within the vehicle Electronic Control Unit (ECU). Comprehensive experimental validation is performed in various track scenarios involving multiple riders. The commercial controller effectiveness is conĄrmed through data comparisons, rider feedback, and safety and performance metrics. The NMPC controller, demonstrating its effectiveness in wheelie prevention, control, and comfort, is tested in straight running conditions. Future work includes adaptive schemes for the commercial controller to account for online variations in rider characteristics and riding style. Additionally, research on a road-slope estimator aims to extend the applicability of the NMPC controller to general road conditions.