Design Study of the Low-Cost Advance Rider Assistance System

Abstract

This work describes the design and implementation of a low-cost Advance Rider Assistance System (ARAS). Motorcycle riders are more prone to the injury during an accident than passengers of the car. For riders those accidents often end up tragically and additionally, there is a higher chance that rider will be involved in the accident than the passenger of a car. Therefore, there is a need for devices that can increase the passive and active safety of bikers. The work describes the design verification and implementation of a simple and affordable assistance system with traffic sign recognition, pedestrian recognition and proximity alert function. Device contains sensory unit equipped with a camera for pedestrian and traffic sign recognition, infrared (IR) rangefinder for proximity measurement and a combination of Global Position System (GPS) sensor and Inertial Measurement Unit (IMU) for the independent speed measurement of the motorcycle. Displaying unit contains Head-Up Display (HUD) and is placed on the helmet. Methodology part describes considered scenarios which could be prevented and possible solutions. In addition to the mentioned functions, the possibility of future extension with smart infrastructure communication functions like Vehicle-To-Vehicle (V2V) and Vehicle-To-Infrastructure (V2I) is taken into account. Based on these considerations, a suitable mechanical solution and used hardware was selected. Design study describes mechanical and mechatronic design and is supplemented by analyses. Implementation part describes software solution (both of sensory and displaying unit) and prototype manufacturing using 3D printing. Test part describes conducted tests and their results, with special emphasis on proximity alert response rate and capability of vision system using cascade classifier. Several further improvements (features which are currently under development, such as night vision, improved scene recognition, people on wheelchairs recognition, etc.) are described at the end of the article. Conclusion involves further work and new ideas that came up during the process.