IN Brief:
- Three partners will test a fleet of three autonomous vehicles at a government research facility.
- The trial will assess following, merging, obstacle avoidance, stopping accuracy and different lane widths.
- Passenger-car-based test vehicles are being used; a commercial freight system has not been deployed.
TIER IV, Taisei Corporation and Taisei Rotec have been selected for a Japanese government demonstration of automated cargo movement on dedicated logistics roads. The partners will operate three experimental autonomous vehicles during an eight-day programme from late October to early November, testing whether coordinated vehicle movements can be supported with simpler onboard sensing and roadside equipment.
The work forms part of the Ministry of Land, Infrastructure, Transport and Tourism’s Autoflow Road initiative, which envisages cargo travelling through purpose-built logistics lanes using unmanned transport equipment. The latest programme concerns the technical requirements for such a system rather than the construction or commercial operation of a freight corridor. Tests will take place at the ministry’s National Institute for Land and Infrastructure Management, using a full-scale tunnel facility and sections of its test track.
Vehicles will run on straight and curved sections to examine the manoeuvres required when several autonomous units share a restricted route. The programme covers leading and following, controlled stopping, travel in opposing directions, parallel running, passing stationary equipment, merging and diverging. It will also test avoidance and overtaking when an object is present on the route, together with stopping accuracy at a designated location. Engineers will assess standard, intermediate and minimum operating widths for the dedicated transport space.
Dedicated logistics lanes can impose more predictable movement patterns than public roads, potentially reducing the number of sensors or fixed installations needed for basic navigation. Vehicles still have to determine their position, maintain separation, respond to obstacles and coordinate encounters where lanes join or narrow. The trial will assess whether those functions can be delivered reliably with equipment selected for the controlled environment, rather than assuming that separation from ordinary road traffic removes the need for safety controls.
TIER IV will provide the experimental vehicles and its Pilot.Auto autonomous driving system, built on the open-source Autoware software. The company is also responsible for vehicle setup, test operations and the collection and analysis of driving data. Taisei Corporation will coordinate the demonstration, evaluate its T-Localizer positioning support technology and investigate equipment simplification. Taisei Rotec will plan and install reflectors, undertake measurements and support work at the test site.
Taisei’s T-Localizer system is intended to support positioning where conventional satellite navigation is unavailable or unreliable. Tunnels impose that constraint directly, so position estimates must come from the vehicle and its surroundings rather than an uninterrupted GPS signal. Taisei Rotec will plan and install reflectors, which can provide reference points for the demonstration, and measure their performance. The partners will investigate whether these arrangements can support reliable movement with a simpler combination of sensors and roadside installations.
The programme follows work undertaken by the same companies in the ministry’s 2025 financial year. That earlier activity examined an autonomous vehicle operating with T-Localizer and Taisei’s T-iPower Road dynamic wireless power transfer technology. The partners said those tests indicated potential applicability to an automated logistics road while identifying questions requiring further examination. The 2026 programme broadens the focus from the operation of an individual vehicle to interaction among several units and the infrastructure necessary to manage them.
The three vehicles scheduled for the latest trial are based on passenger cars. They are experimental platforms and are not the dedicated cargo-carrying equipment that would be used in an eventual Autoflow Road deployment. Results therefore cannot be treated as proof of freight payload capacity, terminal throughput or long-distance commercial availability. Instead, the demonstration will generate evidence about safe movements and the minimum practical supporting infrastructure within the defined test environment.
The demonstration will compare vehicle behaviour in lanes of differing widths. Following an obstruction, making a controlled stop and passing another vehicle impose different demands on position accuracy and lateral clearance. By recording those manoeuvres in standard, intermediate and minimum operating widths, the partners can assess where the available space constrains the operating design. No target commercial speed, freight throughput or financial case has been published, and results from the passenger-car-based platforms should not be presented as capacity estimates for a future cargo network.
The eight-day test is scheduled for late October to early November 2026. Its reported measurements will concern multi-vehicle movements, localisation and the equipment needed along dedicated routes. A subsequent decision to install freight-carrying vehicles or build a commercial corridor would require separate evidence on payloads, operating speeds, maintenance and how the system connects with cargo terminals. None of those later deployment milestones has been announced.
