應用相對加速度資訊於碰撞時間之估測

Time-to-collision Estimation Using Relative Acceleration Information

潘弘裕(Hong-Yu Pan)

指導教授 ： 陳柏全

國立臺北科技大學/機電學院/車輛工程系所/碩士(2008年)

https://doi.org/10.6841/NTUT.2008.00364

摘要

碰撞時間可以做為前方防撞警示系統設計的依據，先前的研究大多是用相對距離除以相對速度來計算碰撞時間，由於忽略相對加速度的影響，其估算上容易產生誤差。本論文考慮相對加速度的影響，提出兩個方法來進行碰撞時間之估測。第一個方法是利用遞迴最小平方法計算相對距離二次曲線的係數，並依此計算出相對距離為零的時間，也就是碰撞時間。第二個方法是利用卡爾曼濾波器預估相對距離、相對速度與相對加速度資訊，以運動學關係計算出預估的碰撞時間。經由模擬結果可得知本論文在相對加速度存在的情況下，相較於未考慮相對加速度影響的演算法，可以有較佳的預估準確性。前方防撞警示系統需能對車輛行經路徑的區域進行偵測，因此本論文中利用車速及偏航角速度計算出所感興趣的區域，再由此區域中選出合適的相對距離資料點，同時也加入跳動點修正策略，以避免因路面顛簸所造成的相對距離資料不連續性。在實車驗證時，本論文所提出的方法比傳統的方法有較佳的估測準確性。

關鍵字

前方防撞警示 ； 碰撞時間 ； 相對加速度 ； 遞迴最小平方法

並列摘要

In order to reduce traffic accidents, Time-to-collision (TTC) are often designed for Forward Collision Warning Systems (FCWS). In the previous studies, usually using relative distance divided by relative velocity to obtain the TTC. Because the result that didn’t consider relative acceleration is incorrect, therefore we propose two methods that using Recursive Least-Squares (RLS) and Kalman filter to determine TTC. In this paper, we proposed to use RLS and Kalman filter algorithms for improving TTC estimation. In RLS algorithm, we used the relative distance as the input item and then used RLS algorithm to get the parameters of a second-order kinematics’ equation based on constant acceleration assumption. In the other way, we used Kalman filter to calculate relative distance, relative velocity and relative acceleration, then used these information to determine TTC. Finally, by way of our simulations show that the proposed TTC is closer to ideal TTC. On the other hand, in order to detect target vehicle correctly in curve, we applied curve detection maneuver and relative distance decision function to obtain the distance to determine TTC.

並列關鍵字

Forward Collision Warning ； Time to-Collision ； Recursive Least-Squares ； relative acceleration.

參考文獻

[4] X. Wang, J. Lins, C.Y. Chan, S. Johnston, K. Zhou, A. Steinfeld, M. Hanson, W.B. Zhang, "Development of Requirement Specifications for Transit Frontal Collision Warning System," California PATH Research Report, ISSN 1055-1425, 2003.

[8] Hirofumi Watanabe, Satoru Kondo, Kikuo Hirano, "Introduction to Suzuki ASV Technologies," Intelligent Vehicles Symposium, Japan, 1996, pp.219-223.

[9] K. Yi, M. Woo, S. H. Kim, S. C. Lee, "An Experimental Investigation of a CW/CA System for Automobiles Using Hardware-in-the-loop Simulations," American Control Conference, USA, 1999, pp.724-728.

[10] A. P. Wang, J. C. Cheng, P.L. Hsu, "Intelligent CAN-based Automotive Collision Avoidance Warning System, "IEEE International Conference on Networking, Sensing and Control 1, pp. 146-151, 2004.

[11] M. Hirano, "Development of Vehicle-following Distance Warning System for Trucks and Buses," Vehicle Navigation and Information Systems Conference, Canada, 1993, pp.513-516.