Mobility of people and goods is a basic economic activity in human society. In the aging society over the world, it is hard to maintain the activity due to the difficulty of walking and driving for elderly people and the labor shortage in delivery, logistics, and transportation. Just delivery services via a digital platform using smartphones could not be a fundamental solution because `physical’ transportation for them still relies on human workers. Autonomous driving of vehicles for transportation, micro-mobility vehicles, and delivery robots could be a solution for mobility in the society. However, ‘standalone’ autonomous driving has drawbacks, particularly in coping with drastic changes in the environment, as it relies on onboard sensors with limited range. Additionally, there are issues with computation and energy capacities. The speaker has proposed a digital-twin technology for mobility with 3D sensor network using multiple light-detection-and-ranging (LiDAR) sensors. It provides environmental information, including the positions of pedestrians and vehicles, to autonomous vehicles and robots. Furthermore, it allows vehicles and robots to ‘outsource’ sensing and computing tasks to it. Sensing and processing necessary for self-driving are performed in the infrastructure sensors and computers instead of onboard ones. The speaker has implemented and demonstrated this technology in real developing smart cities in Japan, which will be presented in the talk.
