KANAZAWA UNIVERSITY
ADVANCED MOBILITY
RESEARCH INSTITUTE

Aiming to realize so-called autonomous driving of “door-to-door” in urban areas, we have been conducting comprehensive research from an academic perspective on the technologies required for autonomous driving. The research area ranges from basic research such as map generation and maintenance required for autonomous driving to research on recognition, decision-making, and control, which are the core technologies of autonomous driving. We also implement a wide range of technologies required for autonomous driving such as research on high-speed computation using large-scale integrated circuits (VLSI), research on safe control of vehicles even on slippery roads such as wet roads and snowy roads, etc.

To develop driving intelligence to enable autonomous driving, it is essential to install equipment such as in-vehicle sensors and computers to mimic the actions of drivers, such as recognition, decision-making , and control. High-precision maps are also required for the realization of more natural autonomous driving. For this reason, the key is using artificial intelligence technologies to handle advanced cognitive and decision-making functions, along with probabilistic robotics to analyze situations robustly in consideration of sensing errors. Since around 1998, we have been conducting research on technologies for driving in urban areas by fusing massive amounts of sensor data for robust recognition and decision-making complicated driving environments, and for the auto-generation of high-precision maps. We have also been researching VLSI algorithms for high-speed advanced image processing.

To enable stable autonomous driving even on slippery roads that make vehicles unstable, we need technologies capable of detecting road surface conditions and enabling proper reactions. Hence, we have focused on tires, the only contact point between the car and the road surface. We have been researching a system to measure the coefficient of friction of the road surface based on tire sidewall and wheel information. This makes it possible to measure the road surface friction coefficient for actual vehicles using tires commonly available in the market. The research results are very valuable, and the technology is expected to be applied not only to autonomous driving in the future, but also to contribute to further developments in existing advanced driver assistance systems such as Anti-lock Braking System(ABS) and Automatic Emergency Braking system(AEB).

Some of these research results were put into practice in the first public road driving demonstration experiment in an urban area, which has been conducted by Kanazawa University since the beginning of 2015. Autonomous driving is being tested in a wide range of traffic environments, from rural areas to metropolitan ones, and from normal to extreme weather in places such as Ishikawa, Hokkaido, and Tokyo. These efforts have been covered by the media in Japan and overseas and has led to further support for a large number of nationwide collaborative research studies among private firms, universities, and governmental offices. Through all of these efforts, we at the Advanced Mobility Research Institute aim to contribute to the research and development of autonomous driving technologies in Japan from an academic perspective.