In recent years there's been an increase in the number or users of micromobility vehicles for everyday commute inside the city. At the same time technologies for autonomous and assisted driving for cars have been improving and entering the market. In this thesis we propose a proof of concept to join both fields and develop an automatic detector using a Convolutional Neural Network for road markings and traffic signs pertaining to micromobility using a relatively small network and a simple tracker we achieve good results. This projects also introduces a new database of traffic signs and road markings.

There is a need of increasing the safety of mobility, considering any type of vehicle, but specially facing the growing use of micro-mobility vehicles, such as scooters, bikes or small cars in urban areas. A fundamental aspect is related to the horizontal signs (signs on the pavement) and hazards encountered in street-lanes. In the last years, vehicle technology has started incorporating different types of sensors to increase safety. These sensors could be used to monitor the condition of the roads. In this project, we propose a system to reinforce safety by using affordab