Computer science that transforms mobility society.

Keywords: Autonomous driving platform, embedded software, real-time systems, IoT, model-based development

A transformation of the mobility society is about to occur, centered around fully autonomous vehicles. Fully autonomous vehicles operate by having computers make decisions regarding steering, acceleration, and braking, which were previously performed by drivers. The computers utilize information from cameras, omnidirectional distance sensors, GPS, and other sources to make these decisions. They manage this information in a three-dimensional space (see diagram) while driving. For the computer to make accurate judgments, it needs to perform various processes simultaneously, as illustrated in the diagram. For example, to drive safely, it requires functions such as object recognition to identify surrounding vehicles and pedestrians, image processing technology to recognize traffic signals, route planning to determine the intended path of the vehicle, and path tracking to ensure it follows the designated route. These functions operate concurrently and are controlled by the computer. In our research lab, we study the mechanisms of computers that effectively manage these various functions simultaneously (completing processing within a specified time). Furthermore, we aim to enhance the safety of autonomous driving through recognition utilizing infrastructure sensors installed at intersections and traffic lights, as well as simulations using digital twin environments.

simulator