An Ontology-Based Recommendation System for ADAS Design

摘要

Advanced driver-assistance systems (ADAS) are an important component in a vehicle for these systems to actively improve driving safety. Although technical development for ADAS is already mature, there are still a few areas that can be improved. In particular, the design of newer ADASs are mainly based on the experience and imagination of car designers and their tests are usually based on hypothetical situations and field models. Without user experience data, it is difficult for car makers to refine and improve their ADAS designs effectively and systematically. In order to help designers optimize their designs and shorten design cycles, a framework of collaborative filtering recommendation is proposed, in which domain ontologies, text mining, and machine learning techniques are used to produce multimedia summaries from data repositories for queries of real accidents, or design issues. The recommendation system framework aims to help designers and car makers improve their efficiency and produce safer vehicles. The data and knowledge bases constructed can be used as the basis of tutorial programs for new entrants and the car makers can reduce their managerial cost and manpower needs.