VLM-Based Advanced Rider Assistance System for Motorcycle Safety
This work addresses the underdeveloped area of motorcycle safety assistance by providing a VLM-based system that improves hazard perception and risk-aware planning for riders.
The paper introduces a novel Advanced Rider Assistance System (ARAS) for motorcycles that uses Vision-Language Models (VLMs) for semantic hazard reasoning and segmentation-based detection to create dense risk maps, which guide a sampling-based planner to reduce hazard exposure. In CARLA simulations, the system achieves higher success rates and lower hazard exposure compared to baselines.
Motorcycles face disproportionately high crash risks compared to cars due to limited protection and heightened sensitivity to surface hazards, yet Advanced Rider Assistance Systems (ARAS) remain underdeveloped relative to Advanced Driver Assistance Systems (ADAS). We propose a novel ARAS that enhances motorcycle safety through semantic perception and risk-aware planning. Our approach leverages Vision-Language Models (VLMs) for contextual hazard reasoning and integrates them with segmentation-based detection to construct dense risk maps. These maps encode both semantic characteristics (e.g., pothole severity, puddle slipperiness) and physical attributes (e.g., size, depth), which produce per-pixel hazard costs that capture motorcycle-specific risks. These maps are used by a sampling-based planner tailored to motorcycle dynamics to recommend throttle and steering actions that minimize hazard exposure while advancing toward the destination. We evaluate our system in different scenarios in the CARLA simulator. Compared to the baseline method, our method achieves higher success rates and lower hazard exposure, while qualitative results demonstrate interpretable risk maps and safe trajectory recommendations.