This article proposes a computational system to assist wheelchair users, improving accessibility through ubiquitous computing technologies. TrailCare uses indoor and outdoor location information to assist wheelchair users, recording their trails and providing context-aware assistance. Trails are historical records of users’ displacements that can be used to develop strategic accessibility solutions, such as security management and recommendation through inferences on behavior. TrailCare contributions are the indoor/outdoor trail-aware strategy and its application to recommend contextualized accessibility resources. The system was implemented and integrated with a motorized wheelchair manufactured by a Brazilian company. The prototype is a complete and functional system based on one of the most used wheelchair models in Brazil. TrailCare was assessed through three practical experiments involving scenarios in a university campus. The first two experiments aimed to evaluate the system functionalities. They consisted of two scenarios that tested practical situations supported by the technologies of context awareness and indoor/outdoor trail awareness. The third experiment focused on evaluating the user experience with the system. It comprised a scenario that was followed by 10 wheelchair users, who were observed by researchers regarding usability aspects. The users also filled out a survey based on the Technology Acceptance Model (TAM). The survey was composed by 10 sentences and the results of each one are discussed. The experiments allowed us to learn 10 relevant lessons about technological and usability aspects of the TrailCare that are recorded in this article. The results also showed 96% of acceptance regarding perceived ease of use and 98% in perceived usefulness. The results of experiments showed the potential for implementing TrailCare in real-life situations, allowing the use of ubiquitous technologies to support accessibility for wheelchair users.

International Journal of Human-Computer Studies