· Research & technology· 日本語で読む
NTT Mobility to Test HD Map-Independent Autonomous Bus on Public Roads in February 2027
response.jp reported on Nov. 15, 2026, that NTT Mobility will test an autonomous bus that does not rely on HD maps on public roads in February 2027.
What happened
response.jp reported on Nov. 15, 2026, that NTT Mobility will test an autonomous bus that does not rely on HD maps on public roads in February 2027. We have not been able to verify the primary source, so please refer to the original report for details.
Why it matters
NTT Mobility’s plan will test on public roads whether the bus can be deployed on new routes without HD maps being prepared in advance. The demonstration will use Level 2 automation, with a driver on board. Before considering commercial operations, performance in Japan’s traffic environment, at night and in bad weather, and coordination with operations management will also need to be evaluated.
Analysis
The plan will test how far route deployment procedures that have relied on preparing HD maps in advance can be replaced by perception and decision-making centered on camera images. NTT Mobility says the demonstration will assess whether the system can be deployed on new routes without HD maps being prepared in advance. The start of the demonstration will therefore verify its potential applicability; it does not mean that map-free operations or commercial use have been achieved. The technical setup also differs from a simple “leave it to AI” approach. According to the report, it is an E2E system that integrates surrounding-environment perception, driving plans and vehicle control, while functions such as perception and decision-making are divided, and rules covering stop signs, right of way and speed limits are also incorporated. The system is said to allow improvements for specific challenges, such as detecting traffic lights in Japan. At the same time, whether AI decisions can be explained and verified, and how much performance depends on training data, are also matters for evaluation. The implementation focus will be not only the advantages of integration, but also how much work remains for function-by-function verification and regional adaptation. The public-road demonstration will begin at Level 2, with a driver on board. Evaluation will cover adaptation to left-hand traffic, traffic lights and regulations, as well as responses to illegally parked vehicles, pedestrians, cyclists, nighttime conditions and bad weather. This scope expands the assessment of map-free operation beyond the effort involved in setting routes to include perception and driving quality in actual road environments. However, the report does not provide benchmarks for comparing driving records or performance, or specific figures for operating costs. NTT Mobility has also been associated with a public-road driving plan for an e-Palette with May Mobility, reported in May Mobility and NTT Mobility plan Japan’s first public-road drive of an autonomous e-Palette. The vehicle in the current plan is made by eVersum, and the system by Imagry, a different combination from the earlier report. There is no basis to treat the two as the same deployment approach, but they provide grounds to assess NTT Mobility’s public-road efforts across more than one vehicle and system configuration. The next points to watch are how much adjustment will still be needed to deploy the system on new routes, and how safety evaluations and coordination with remote monitoring and operations management will be tied to the requirements for commercial service.
Sources
Translated from the Japanese original: /news/ntt-mobiritei-hd-mappu-ni-izon-shinai-jidouunten-basu-wo-202-f6e4ac. The Japanese version is authoritative if they differ.