Description of the Japanese Computer-controlled Vehicle System (CVS)

This description has been derived largely from a report entitled Automated Guideway Transit: An Assessment of PRT and Other New Systems, U.S. Congress, Office of Technology Assessment, June, 1975

The Ministry of International Trade and Industry (MITI) sponsored development of a very sophisticated PRT system, called CVS (Computer-controlled Vehicle System in the early 1970's. In addition, a small dualmode vehicle system, also called CVS, was demonstrated at an International Ocean Exposition on Okinawa in July 1975.

Planning for the CVS PRT system development work began in Japan in 1968. At that time preparations were made for a "traffic game" to be demonstrated at the Osaka World Exposition, which was held from March to September 1970. The demonstration in the Automobile Industries Pavilion consisted of more than ten specially designed electric vehicles operating individually under computer control on a checkerboard-like guideway network with intersections every five meters (16.4 feet). The two-seat vehicles communicated with the central computer through an underground communication channel.

Though primarily designed as an exhibition facility, this demonstration accomplished several things, including:

-- Development of elementary computer logic for controlling a small fleet of vehicles

-- Development of techniques for managing vehicle at-grade crossings on a highly integrated network of intersections

-- Assessment of the use of small automated vehicles in public service

A more extensive concept for CVS was formulated in July 1970. In the autumn of 1970 work on the basic design of this system began with support of the Ministry of International Trade and Industry (MITI). Miniature models of vehicles and a guideway were constructed. A total system with 1,000 vehicles was simulated on a large computer as the basis for preparing fundamental technical specifications.

Based on this research, a reduced-scale experiment was prepared from April to October 1971. Cars were developed for a network representing the central 300-meter square area (984 feet) of the Ginza District in Tokyo. CVS cars at 1:20 scale were operated under computer control for the public at the 18th Tokyo Motor Show from 28 October to 21 November 1971. Though the vehicles and guideway were one-twentieth scale, the computer-control system was full scale. Thus, the experiment provided an opportunity to exercise both the computer hardware and software for an extensive automated network system.

At the conclusion of the Tokyo Motor Show in November 1971 full-scale development of CVS began. MITI financed construction of test facilities on the site of Japan's first automobile test track at Higashimurayama, about 30 km (18 miles) west of Tokyo. The test track configuration is shown in the following diagram. See an aerial view of the CVS test facilities and a diagram of the test track configuration at Higashimurayama, Tokyo, Japan. The maintenance area and an off-line station are shown within the 100meter grid in the middle of the test loop.

The total length of the guideway was 4.8 km (2.9 miles). At the top of the diagram, 2 km of straight track permit high speed operations at 60-80 km/hr (36-48 mph). Two parallel traffic lanes at the bottom of the diagram permit high-speed lane changing experiments. The diamond-shaped portion in the center represents the grid in a lowspeed network. One side of the grid is 100 meters (328 feet) long, which is technically the minimum distance between stations. The test track is designed with two at-grade crossings to check performance of the vehicle control system at these intersections. The telephone-shaped track inside the grid has a circular guideway with a radius of 5 meters (16.4 feet) at both ends and is used as a maintenance area. The control center, vehicle storage yard and passenger cargo station are below the maintenance track. A second passenger/cargo station is located at midpoint on the upper high-speed track.

Control System

The CVS system is controlled by a synchronous moving block system using three separate computer systems. The first one is the Hitachi computer system, which controls the vehicles high speed operation, i.e., the outer-ring (speed between 40- 60 km/hr) ; the second is the Toshiba computer system controlling the low speed operation, i.e., the inner-ring (speeds below 40 km/hr) and moving blocks on the guideway; the third one is the Fujitsu computer system which functions as the supervisory computer, controlling overall system operation, and monitoring the other two computer systems.


While originally planned for a 100-vehicle test operation, inflation reduced the scope to 60 vehicles, as follows:

Phase 1 Phase 2
"All up" vehicles:

Passenger 0 13
Cargo 0 6
"Test bed" chassis:

Passenger 16 27
Cargo 8 14
TOTAL 24 60

All vehicles were delivered to the site, and test operations were scheduled over two phases as discussed in a following section. A variety of vehicle configurations were available for testing. The "all-up" versions were complete prototypes, whereas the "test-bed" chassis contain only the essential propulsion, control and braking equipment. Passenger vehicles generally have four seats--facing backwards for safety purposes. Standing was not permitted on automated transit systems at this time in Japan. Other passenger-carrying versions have forward facing seats; two of the seats can be folded to provide space for a baby carriage or hand baggage. See a photograph of a cargo and passenger version of the CVS vehicle, manufactured by the Toyo Kogyo Co., Ltd.

The cargo vehicles are designed with a capacity of 300 to 400 kg (660-880 pounds), and since the vehicles have no springs, they remain level with the cargo platform. Three versions have been observed: a flat-bed type with two conveyor belts built into the floor, a panel truck type and a postal truck type.

System Characteristics

The following is a summary of the main CVS characterislics:

Guideway Steel "I"-beams for the running surface

Steel sections for the guide groove containing controls and power rails

Dimensions: 2m (6.6 ft) wide, 0.8m (2.6 ft) deep

Span: 20-30m (66-98 ft)

Maximum Grade: 10 percent

Minimum Radius: 5m (16.4 ft)

Stations Passenger or cargo


Vehicles 4 passengers (all seated)

Pneumatic rubber tires

Propulsion: 200 v AC motor

Speed Normal: 40 km/hr (24 mph)

High speed: 60 km/hr (36 mph)

Maximum: 80 km/hr (48 mph)

Headways 1 second
Braking Electric regenerative for high speeds - 0.2 G

Friction for low speeds - 0.2 G and 0.5 G

Emergency, explosive activitated - 2.0 G

Operations Per lane: 3,600 veh/hr, 14,400 seats/hr

Entrained operations are contemplated with 20 to 30 passengers/train

Program Schedule

The Higashimurayama project begain in 1971 and its basic design was completed by the middle of 1972. In the autumn of the same year the maintenance guideway and the first experimental vehicle were completed. Basic driving tests under manual control commenced shortly thereafter. In the spring of 1973 basic experiments with computer control began. The full length of guideway was constructed in the autumn of 1973. At this time the second stage of experiments began, including: computer control of several vehicles, operations for passenger service at stations, control of automatic loading and unloading of freight containers, lane changing experiments and overtaking of vehicles at high speed.

Phase I of the Higashimurayama project was extended through the spring of 1976. The objective during this phase was to accumulate itemized basic experimental data. Condensed experiments were conducted to maximize the capability of vehicles to follow the guide target. The system reliability including the reliability of mobile communications between the vehicle and the wayside computer. Phase I was intended to prove the technical practicability of CVS through item tests of components and subsystems.

Speed tests were conducted at 60 km/hr (36 mph) and headways of five seconds have been consistently achieved at 20 to 30 km/hr (12 to 18 mph). No attempt has been made to test operations at headways close to one second. To reduce risks to equipment and vehicles, the short-headway tests were not attempted until they could be carried out safely, sometime during Phase II.

The second phase lasted several years, starting in the spring of 1976. The purpose of Phase II was to conduct a total trial of the the system, with 60 vehicles operating on the test track under complete computer control. This phase was primarily concerned with cargo operations and did not emphasize passenger transport. Reliable and consistent one-second headways were attempted with those vehicles designed for such operations. Phase II aso incorporated mechanisms for collision avoidance, obstacle detection/avoidance, and a 0.5 G emergency braking system.


The test facilities and scope of the experimental operations are impressive

The three-tiered computer control system seems unnecessairly complex. The Japanese claim it was no more difficult to design the computer interfaces than a single computer system to do all the necessary functions

Use of an explosive-actuated emergency brake to avoid a catastrophic collision would produce a 2.0 G deceleration force. While passengers face backwards on high-backed seats to minimize the effect of such an instantaneous stop, there is concern that injuries may still occur.

Project Management

The CVS Project was under the general direction of MITI and is sponsored by the Japan Society for the Promotion of Machine Industry (JSPMI). The Ministry spent approximately $10 million on the project, mostly for the test track and related facilities. JSPMI channels funds derived from other industry associations to help fund the project. A consortium of eight private industrial firms shared in the cost and provided technical resources. These firms were: Toyo Kogyo, Mitsubishi Jukogyo, Tokyo Shibaura Denki, Hitachi Seisakusho, Fujitisu, Sumitomo Denki Kogyo, Nippon Denki and Shin-Nihon Seitetsu. The CVS Project was managed by a team from Tokyo University and MITI.

home2.gif (1492 bytes)

Last modified: June 30, 2008