This paper describes a method for categorizing road roughness into three primary groups, ranging from low to high roughness, utilizing an accelerometer and GPS that can be mounted on a wheelchair. Additionally, the system is designed to autonomously detect the presence of potholes. The project aims to facilitate the development of a system enabling wheelchair users to identify routes with minimal roughness, thereby optimizing comfort during navigation. The system is conceptualized within an edge computing framework, where a server conducts computations and a smartphone application presents the results and recommends the most favourable path.