臺灣博碩士論文加值系統

隨著汽車的改善及其便利性，汽車已成為現代人最重要的交通工具
。隨著汽車數量的增加，車禍事故也相形提高。車禍常常會使人受傷或造成嚴重的傷害，且行人死亡在因車禍而喪生數目中佔有很大比例，因此有必要改善行人安全。行人死亡在交通意外事故死亡中，佔有相當大的比例。研究指出，50%以上的行人死亡是由於頭部傷害所造成，而傷害多因頭部撞擊至車體的引擎蓋與擋風玻璃處。另外國內在肇事現場資料蒐集與作業流程，皆以人工的方式進行記錄及量測，常造成現場量測錯誤及資料蒐集不完整的問題。本研究採用PC-Crash 3D模擬軟體來分析行人在交通意外事故中之頭部受傷指數，以提供法院及各機關在判定責任歸屬時之參考。首先探討的為車輛型式，該部分使用轎車及巴士，並做行人之運動及動態分析。基於這些結果，探討行人頭部傷害指數。因碰撞時車輛與行人各部位間會彼此影響，因此再度探討車輛參數及行人參數對行人安全之相關性，車輛參數包含車種、車重、車速，行人參數包含身高、體重。最後經由徑向基類神經網路來訓練車輛與行人參數，希望在經由訓練後，受訓練和期望的頭部受傷指數(HIC)值愈接近愈好，以提高肇事鑑定之準確率。整體而言，肇事重建軟體之運用，有助於提升鑑定品質與提高鑑定之公信力，並協助降低或減輕行人傷害及社會成本。

As automobile improved, they have become one important transportation means for modern people. While the use of automobile increases, the accidents happen more frequently. Automobile accidents normally would cause injuries to humans. Among accidental causalities due to vehicle crash, pedestrian fatalities constitute a large portion. Therefore, there is a need to improve pedestrian safety. Previous studies show that more than 50% of pedestrian fatalities are caused by head injuries, most of which are due to impact with the hood or windshield. Besides, in our country we record and measure the accident data in an artificial way, which usually leads to problems of incorrect measurement and incomplete data collection. This study uses the PC-Crash simulation software to analyze the Head Injury Criterion of the pedestrians in traffic accidents. These accidents reconstruction data can thus be applied by the authorities concerned as the major reference. The first part of the study explores vehicle type’s influence on pedestrian safety. Two major types of vehicle, car and bus, are used. The responses in kinematics and dynamics are described. Based on the results, the possible head injuries of the pedestrians are discussed. Since the nature of direct impact, the interacting properties between pedestrian segments and vehicle components are central to pedestrian impact, the second part of the study is to discuss pedestrian safety’s relevance with vehicle parameter and pedestrian parameter. Vehicle parameter includes type, mass and velocity. Pedestrian parameter includes height and weight. The study uses RBF to train the vehicle and pedestrian parameters. After training, the closer the trained output HIC is to the expected value, the better the performance is. Overall, using PC-Crash software will increase the quality and credibility of accident authentication and helps in reducing head injuries and social cost.

摘要.....................................................I
Abstract................................................II
誌謝 ....................................................IV
Contents.................................................V
List of tables.........................................VII
List of figures.......................................VIII
Chapter 1 Introduction...................................1
Chapter 2 Theory.........................................4
2.1 Introduction of PC-Crash.............................4
2.1.1 The Trajectory Model...............................6
2.1.2 The Multibody Model ..............................13
2.2 Exploration of Head Injury Criterion................23
2.2.1 Background on Head Injuries ......................23
2.2.2 Head Injury Tolerance ............................25
2.2.3 Head injury criterion functional..................26
2.2.4 Pedestrian Accident Kinematics ...................29
2.3 Radial Basis Function Neural Network................31
2.3.1 RBF network architecture..........................32
2.3.2 Training of RBF networks..........................34
Chapter 3 Working with PC-Crash.........................36
3.1 Loading Vehicle.....................................36
3.2 Setting Vehicle.....................................38
3.3 Environment of the accident.........................40
3.3.1 Scenes............................................40
3.3.2 Placement of vehicles, pedestrians and obstacles..41
3.3.3 Sequences.........................................42
3.4 The Pedestrian Model................................43
3.5 Paths...............................................45
Chapter 4 Analysis of Simulation Results................46
4.1 Head Injury of Pedestrians with the Height..........46
4.2 Summary.............................................47
Chapter 5 Conclusion....................................49
References..............................................50
Appendix................................................52

List of tables
Table 1.1 Head Injury Criterion for Various Dummy Sizes.52
Table 2.1 Vehicle Values................................52
Table 2.2 Multibody system..............................53
Table 4.1 Comparative predictions of three models.......53

List of figures
Figure 2.1 the three coordinate systems.................54
Figure 2.2 the vehicle model rotations..................54
Figure 2.3 the pedestrian multibody model...............55
Figure 2.4 inertial space and body fixed coordinate systems.................................................55
Figure 2.5 steps during the simulation of the multibody system..................................................56
Figure 2.6 joint definition.............................56
Figure 2.7 ellipsoid to ellipsoid contact...............57
Figure 2.8 ellipsoid to plane contact. .................57
Figure 2.9 Head section (left) and fatal head injury with subdural hematoma(right)................................58
Figure 2.10 HIC-determining maximum-intervals...........58
Figure 2.11 the crash performed without an airbag.......59
Figure 2.12 the crash performed with an airbag..........59
Figure 2.13 Structure of a single-hidden-layer RBFN with n input values, m basis functions and one output value....60
Figure 3.1 Vehicle database.............................61
Figure 3.2 load a custom vehicle........................61
Figure 3.3 Vehicle Geometry.............................62
Figure 3.4 Vehicle suspension properties................63
Figure 3.5 Occupants and cargo..........................64
Figure 3.6 Rear brake force.............................65
Figure 3.7 Vehicle shape................................66
Figure 3.8 Trailer......................................67
Figure 3.9 Example of a 2D scene drawing................68
Figure 3.10 Draw toolbar................................68
Figure 3.11 3D road object dialog box...................69
Figure 3.12 Example of a 3D scene drawing...............69
Figure 3.13 position & velocity dialog box..............70
Figure 3.14 Sequences dialog box........................71
Figure 3.15 Multibody model dialog box..................72
Figure 3.16 Path & Path anchor..........................72
Figure 4.1 HIC Values at Different Speeds with braking
(Car mass：1000kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................73
Figure 4.2 HIC Values at Different Speeds without braking
(Car mass：1000kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................73
Figure 4.3 HIC Values at Different Speeds with braking
(Car mass：1500kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................74
Figure 4.4 HIC Values at Different Speeds without braking
(Car mass：1500kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................74
Figure 4.5 HIC Values at Different Speeds with braking
(Car mass：2000kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................75
Figure 4.6 HIC Values at Different Speeds without braking
(Car mass：2000kg, Pedestrian height/mass：160cm/40kg, 50kg, 60kg).............................................75
Figure 4.7 HIC Values at Different Speeds with braking
(Bus mass：10t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................76
Figure 4.8 HIC Values at Different Speeds without braking
(Bus mass：10t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................76
Figure 4.9 HIC Values at Different Speeds with braking
(Bus mass：15t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................77
Figure 4.10 HIC Values at Different Speeds without braking
(Bus mass：15t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................77
Figure 4.11 HIC Values at Different Speeds with braking
(Bus mass：20t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................78
Figure 4.12 HIC Values at Different Speeds without braking
(Bus mass：20t, Pedestrian height/mass：160cm/40kg, 50kg, 60kg)...................................................78
Figure 4.13 HIC Values at Different Speeds with braking
(Car mass：1000kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................79
Figure 4.14 HIC Values at Different Speeds without braking
(Car mass：1000kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................79
Figure 4.15 HIC Values at Different Speeds with braking
(Car mass：1500kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................80
Figure 4.16 HIC Values at Different Speeds without braking
(Car mass：1500kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................80
Figure 4.17 HIC Values at Different Speeds with braking
(Car mass：2000kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................81
Figure 4.18 HIC Values at Different Speeds without braking
(Car mass：2000kg, Pedestrian height/mass：170cm/50kg, 60kg, 70kg).............................................81
Figure 4.19 HIC Values at Different Speeds with braking
(Bus mass：10t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................82
Figure 4.20 HIC Values at Different Speeds without braking
(Bus mass：10t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................82
Figure 4.21 HIC Values at Different Speeds with braking
(Bus mass：15t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................83
Figure 4.22 HIC Values at Different Speeds without braking
(Bus mass：15t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................83
Figure 4.23 HIC Values at Different Speeds with braking
(Bus mass：20t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................84
Figure 4.24 HIC Values at Different Speeds without braking
(Bus mass：20t, Pedestrian height/mass：170cm/50kg, 60kg, 70kg)...................................................84
Figure 4.25 HIC Values at Different Speeds with braking
(Car mass：1000kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................85
Figure 4.26 HIC Values at Different Speeds without braking
(Car mass：1000kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................85
Figure 4.27 HIC Values at Different Speeds with braking
(Car mass：1500kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................86
Figure 4.28 HIC Values at Different Speeds without braking
(Car mass：1500kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................86
Figure 4.29 HIC Values at Different Speeds with braking
(Car mass：2000kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................87
Figure 4.30 HIC Values at Different Speeds without braking
(Car mass：2000kg, Pedestrian height/mass：180cm/70kg, 80kg, 90kg).............................................87
Figure 4.31 HIC Values at Different Speeds with braking
(Bus mass：10t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................88
Figure 4.32 HIC Values at Different Speeds without braking
(Bus mass：10t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................88
Figure 4.33 HIC Values at Different Speeds with braking
(Bus mass：15t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................89
Figure 4.34 HIC Values at Different Speeds without braking
(Bus mass：15t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................89
Figure 4.35 HIC Values at Different Speeds with braking
(Bus mass：20t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................90
Figure 4.36 HIC Values at Different Speeds without braking
(Bus mass：20t, Pedestrian height/mass：180cm/70kg, 80kg, 90kg)...................................................90
Figure 4.37 Motion of pedestrian in frontal-side crash..91
Figure 4.38 Comparison of HIC values with speed of the vehicle.................................................91
Figure 4.39 the acceleration values of pedestrian’s head impacting the bus.......................................92
Figure 4.40 the acceleration values of pedestrian’s head impacting the car.......................................92

1.Henn, H.-W. 1998, “Crash Tests and the Head Injury Criterion”, Teaching Mathematics and Its Applications Volume 17, No. 4.
2.Versace, J.“A Review of the Severity Index, Proceedings 15th Stapp Car Crash Conference”, SAE paper 710881, pp 771-796.
3.Steffan D. 2008,“Operating Manual of PC-Crash,MEA”, Austria.
4.Steffan D. 2008,“TechnicalManual of PC-Crash,MEA”, Austria.
5.Moser A., Steffan H., Kasanický G. 1999, “The Pedestrian Model in PC-Crash –The Introduction of a Multi Body System and its Validation” , Society of Automotive Engineers,0445.
6.Deck C., Willinger R. 2009, “Head injury prediction tool for protective systems optimization” , Strasbourg University, IMFS-CNRS.
7.Portions J. 1990,“Automotive Safety: Anatomy, Injury, Testing and Regulation”, SAE.
8.Tou, J. T., Gonzalez, R. C. 1974“Pattern Recognition. Reading, MA: Addison-Wesley”.
9.Park, J., Sandberg, J. W. 1991, “Universal approximation using radial basis functions network” , Neural Computation, vol. 3, pp. 246-257.
10.Poggio, T., Girosi, F. 1990, “Networks for approximation and learning”, Proc. IEEE, vol. 78, no. 9, pp. 1481-1497.
11.Haykin, S. 1994“Neural Networks: A comprehensive Foundation. Upper Saddle River”, NJ: Prentice Hall.
12.Moody, J. 1989, “Fast learning in networks of locally-tuned processing units” , Neural Computation, vol. 1, pp. 281-294.
13.Bors, A.G., Pitas, I. 1996, “Median radial basis functions neural network”, IEEE Trans. on Neural Networks, vol. 7, no. 6, pp. 1351-1364.
14.M.J. Orr. 1996,“Introduction to Radial Basis Function Networks, Centre for Cognitive Science” , Edinburgh University.
15.Zimmermann H.-J. 1995,“Neuro and Fuzzy: Technologies — Anwendungen” , Du¨sseldorf, VDI-Verlag.
16.Patterson D. 1996, “Artificial Neural Networks: Theory and Applications”, Prentice-Hall, Munich, pp. 159–232.
17.Abayomi O., John G., “An Analytical Assessment of Pedestrian Head Impact protection” , United KingdomPaper Number 98-SlO-W-17.
18.George T. 2007, “Aspects Regarding The Impact Speed, AIS and HIC Relationship for Car-Pedestrian Traffic Accidents” ,Fascicle of Management and Technological Engineering, Volume VI (XVI).