4. Lecture 2
Safe Driving Support System Using Inter-vehicle Communications

Akira Iihoshi
Chief Researcher, Automobile R&D Center,
Honda R&D Co., Ltd.

<Abstract>
Safe driving support systems are classified into two types: Autonomous detection-type driving support and communication-type driving support. The latter is further sub-classified into a roadside information driving support system that utilizes roadside information, and an information exchange driving support system that utilizes inter-vehicle communications.

In planning the concept for the information exchange driving support system, we analyzed accidents to create accident models, sampled accident patterns and then analyzed the system requirements for supporting safe driving. Next, communication and position evaluation technologies were examined and the concept specifications were created. Furthermore, we have verified application data specifications, communication range, the number of vehicles within the communication range, radio-wave propagation, drivers' receptivity and application evaluations.

Looking at the latest situation regarding accidents, the number of persons killed is decreasing; however, the number of accidents and number of seriously injured people have not decreased. In reviewing accident patterns, the largest number of accidents involves pedestrians and vehicles, followed by vehicle-to-vehicle accidents (e.g., colliding with each other), head-on collisions and right-hand-turn accidents, in that order. As for the characteristics of these four types of accidents, the results of analysis confirmed that: in the case of making right-hand turns, many accidents involving 50cc motorcycles occurred at intersections; in the case of vehicles colliding with each other, many accidents occurred at small intersections without traffic signals; in the case of pedestrians, many accidents occurred when crossing simple roads not equipped with pedestrian crosswalks; and in the case of head-on collisions, many accidents occurred on simple roads. As delayed recognition, such as failure to confirm safety and failure to pay attention to the road ahead, accounted for 71% of the accidents by factor analysis, the safe driving support system will focus on these factors.

Seven accident patterns (i.e., right-turn, head-on collision, vehicles colliding with each other, pedestrian-related, rear-end collision, left-turn and changing lanes) were used for the ASV (Advanced Safety Vehicle) project. A trial calculation was conducted simulating the installation of an inter-vehicle communications system in all vehicles, and assuming that the system and all devices worked accurately. In such a case, it is expected that there would be a 28% reduction in the number of accidents resulting in death, and a 38% reduction in the number of accidents resulting in serious injury.

The abovementioned simulation assumes that each vehicle can periodically exchange information on its speed, location and so on to other vehicles. In this case, obtaining the information about other vehicles, drivers can more easily recognize the traffic environments around their vehicles. Safe driving is thereby supported through providing information about possible invisible dangers to the driver.

Since there will be few vehicles in which the system is installed when system operation begins, it is believed that the service itself should be limited, thereby urging drivers to maintain alertness at all times. As the share of vehicles using the system approaches 100%, it is expected that higher levels of functionality will be added; with system functions progressing from perceptual functions to providing information for safe driving support and brining attention to dangers and braking assistance.

For communication specifications, it was decided to use the 5.8GHz band to broadcast communications data in packets of 100 bytes. Furthermore, the communication transmission distance is defined based on the idea that the information should be provided in such a way that allows the driver sufficient time to stop the vehicle. The system's target communication range, required data and data format were defined based on the accident pattern analysis. The basic data to be transmitted includes items such as vehicle identification, location, speed, gear position and braking condition. In addition to the data required to support safe driving, entertainment-related information is included to enhance the enjoyment and convenience that inter-vehicle communications provides.

The aim of the experimental evaluation was to verify whether or not information from surrounding vehicles works effectively to support safe driving in close proximity situations that are experienced in actual traffic environments. The two points checked were:
If a vehicle's location could be measured as accurately as desired, and
If the driver could use the system correctly without misunderstanding and/or confusion when in a traffic environment where other vehicles not equipped with the system were on the road as well.

The experiments were conducted from July to October 2005, on a lap-style test course and intersections of straight-road sections at the Civil Engineering Research Institute of Hokkaido, Tomakomai, Hokkaido, Japan. Evaluations for the seven accident types previously mentioned were carried out. The items checked included:
- If the timing of information provided was appropriate.
- If the driver could understand the contents of the information provided.
- If the information was accepted as the actual conditions.
- If the driver could understand what to do with the information provided.
- If providing the information was annoying to the driver.
- If the communications range was too far or too close.
- If the expected functions of the system were realized.

The results of the experiments are as follows:
- The provision of information about other vehicles is effective as a preventive safety measure.
- There are some applications in which the current location accuracy cannot pinpoint vehicle location in a traffic lane.
- The system is effective even when used in an environment where other vehicles not equipped with the system are present.
- A sufficient communications range for inter-vehicle communications is achieved when a relay device is incorporated.
- Communications performance and radio-wave transmission characteristics using the 5.8GHz band were confirmed.

Problems remaining to be resolved include the following:
- Location accuracy is still erroneous in that it is not clear whether a vehicle is passing or not passing.
- There are course prediction errors due to delays in receiving data.
- Confirmation of how many vehicles are able to communicate with each other during inter-vehicle communication.
- Development of a communication system that considers blind-corner communications.
- Implementing applications successfully.
- Designing a HMI (Human-Machine Interface) that enables drivers to understand information instantly and does not interrupt driving.

Finally, we conducted a review of the present situation of the infrastructure and the possibility of cooperation to implement a safe driving support system that realizes the world's safest road traffic society by the year 2010. According to the accident analysis performed, it is important to identify the types of systems that can increase efficiency based on accident pattern data, and to draw scenarios that reduce accidents through the introduction of complementary systems. In other words, a process of dissemination is required: first, communications between the roads and vehicles is introduced, and then inter-vehicle communications will have an increasingly greater effect on decreasing the number of accidents as the number of vehicles using the systems increases.