A multi-layer functional safety framework integrating sensor fusion predictive collision detection and intelligent emergency response for powered two-wheelers in urban traffic

Powered two-wheelers (PTWs) remain one of the most vulnerable modes of urban transportation due to limited structural protection, unstable vehicle dynamics, and increasingly complex mixed-traffic environments. Although several active safety technologies have been proposed, existing solutions largely address isolated safety functions and rarely provide an integrated framework based on established functional safety principles. This study presents a comprehensive multi-layer functional safety architecture for powered two-wheelers by adapting ISO 26,262 and SOTIF methodologies to the unique operational characteristics of motorcycles. The proposed framework integrates multi-sensor perception, Extended Kalman Filter (EKF)-based sensor fusion, Long Short-Term Memory (LSTM)-based collision prediction, rider intervention mechanisms, and automated emergency response into a unified safety ecosystem. Hazard Analysis and Risk Assessment (HARA) is employed to identify safety-critical functions and derive Automotive Safety Integrity Level (ASIL) allocations for key hazards. The framework is evaluated using 1,200 Monte Carlo simulation trials across varying urban traffic densities. Experimental results demonstrate an average collision detection rate of 88%, a 5.4% false alarm rate, a mean 2.19 s time-to-collision warning margin, and an average 32.7% reduction in simulated impact speed, indicating substantial improvements in proactive hazard detection and post-crash response. In addition, the framework’s design considerations address practical deployment factors including feasibility, interoperability, and cost effectiveness. The proposed architecture establishes a structured functional safety methodology specifically tailored for powered two-wheelers and provides a scalable foundation for future AI-enabled rider assistance systems, connected mobility, and intelligent transportation applications.