Mitsubishi tests smart vehicle infrastructure system to boost autonomous bus safety
Press Release, 19 December 2025
Mitsubishi Heavy Industries (MHI) has launched a real-world demonstration of its vehicle-infrastructure integration system to support the safe operation of autonomous buses in Shimotsuke City, Japan. The trial, running from mid-December 2025 to late February 2026, is part of broader efforts to accelerate the adoption of driverless public transport — especially on routes struggling with manpower shortages. This demonstration is conducted in cooperation with the Ministry of Land, Infrastructure, Transport and Tourism (MLIT), Tochigi Prefecture, Shimotsuke City, and local transport operator Kanto Transportation, with MHI responsible for installing the system and collecting data throughout the test period.
The new system uses a combination of ground-mounted cameras and advanced AI image analysis to detect and interpret surrounding road conditions that autonomous buses may otherwise struggle to sense on their own. By analysing camera images in real time, the technology can recognize other vehicles, pedestrians and obstacles, calculate positions, speeds and directions, and then relay that information directly to the bus. This approach helps to cover blind spots like vehicles approaching from behind or at awkward angles — that on-board sensors on driverless vehicles might miss, enhancing safety and smoothness during departures or complex manoeuvres. MHI’s use of relatively affordable monocular optical cameras also aims to keep system costs manageable, supporting wider adoption across different regions.
This initiative builds on MHI’s earlier research into vehicle-infrastructure cooperation for intersections and other challenging driving scenarios, and reflects a broader push to integrate infrastructure-side intelligence into autonomous transport networks. By filling gaps in onboard perception and sharing critical road data in real time, the company hopes to make autonomous buses safer, more reliable and better suited to meet the needs of communities facing driver shortages while maintaining vital public transportation links.
Compiled using AI
