| Project/Area Number |
08650626
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
交通工学・国土計画
|
|---|
| Research Institution | Yamaguchi Univeristy |
|---|
Principal Investigator |
HISAI Mamoru Yamaguchi University, Faculty of Engineering, Associate Professor, 工学部, 助教授 (80110237)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAMURA Youichi Yamaguchi University, Faculty of Engineering, Associate Professor, 工学部, 助教授 (20035075)
|
|---|
| Project Period (FY) |
1996 – 1997
|
| Project Status |
Completed (Fiscal Year 1997)
|
| Budget Amount *help |
¥1,800,000 (Direct Cost: ¥1,800,000)
Fiscal Year 1997: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 1996: ¥1,300,000 (Direct Cost: ¥1,300,000)
|
|---|
| Keywords | signal coordination / human judgment / neural network / ゲ-ミング / 人間の判断力 / 交通応答制御 / 追従現象 |
|---|
| Research Abstract |
In this paper, human ability capable of visually judging the whole traffic situation is introduced into coordinated signal control to improve control efficienyc. For that purpose, a traffic simulation is developed to simulate traffic behavior according to the indication of traffic signals. From gaming simulation where a subject can join the coordinated control by hitting some keys to change or extend any signal indications watching traffic situation on the computer screen, the control strategies suitable to traffic conditions are searched. The control strategies are learned and modeled by neural network. Important research results are as follows :
(1) A computer program to simulate traffic flow at signal coordination was developed to perform gaming simulation.
(2) Some observations were carried out to know behavior of acceleration and deceleration at intersections and propgation behavior of starting shock wave. By using the results, some parameters of simulation program were adjusted.
(3) Gaming simulations to switch or extend signal indications by hitting keys were performed watching traffic situation on the computer screen.
(4) Gaming simulations were repeated to obtain better control strategies than simple progressive control.
(5) The control strategies were leared by neural network to generate control strategies according to given traffic conditions.
(6) From the simulation results applying the control strategies obtained from the model, it was found that although control efficiencys were not always satisfying, some good control strategies were obtained depending on traffic conditions.
|
