Vehicle Dynamics Modeling and Simulation Approach for Safety

Vehicle Dynamics Modeling and Simulation Approach for Safety Performance Evaluation of Cable Barriers in Curves Master’s Thesis Defended by Simon Bonnevie Research Assistant at the Center for Intelligent Research Systems August 23 rd, 2011

Outline • Background and Significance • Vehicle Dynamics Analysis Program 1. 2. 3. 4. Problem Statement Traffic Safety Facts Cable Barriers Underride 1. 2. 3. 4. Simulations in Carsim Output Data Graphic Output Main Program 1. 2. 3. 4. Overview Table Headlines Nomographs Table Analyses • Results • Conclusion 2

Background and Significance • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results Outline • Problem Statement • Traffic Safety Facts • Cable Barriers • Underride • Conclusion 3

Problem Statement • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results • Developing tools and methods to evaluate the safety performances of cable barriers, and optimize guidelines for their design selection and appropriate placement. • Providing insight on the effects of horizontal curvature in the dynamics of roadway departure vehicles. • Conclusion 4

Traffic Safety Facts • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results • Conclusion FARS and FHWA (2009) [38, 15] 5

Cable Barriers • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results • Conclusion 6

Cable Barriers • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results Advantages: • • Low-cost Easy to deploy Low deceleration forces upon impact FHWA review of 25 states (2004): typical effectiveness of 95% [41] Drawbacks: • • • Fewer access points for emergency and maintenance vehicle Increase in reported crashes (because of less recovery area) Need to accommodate for lateral deflection Adverse effects in curves Capturing capability reduced to zero until repair • Conclusion 7

Cable Barriers • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results AASHTO RDG (2006) [6] • Full-scale crash tests standardized in NCHRP Report 350 [42] and MASH [7] • Warrants based on ADT and median width • State-specific warrants established from benefit-to-cost analysis • State-specific warrants established from crash studies • Conclusion 8

Underride • Background and Significance – Problem Statement – Traffic Safety Facts – Cable Barriers – Underride • Vehicle Dynamics Analysis Program • Results • Conclusion 9

Vehicle Dynamics Analysis Program • Background and Significance • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results Outline • Simulations in Carsim • Output Data • Graphic Output • Main Program • Conclusion 10

Simulations in Car. Sim • Background and Significance Run Control Screen • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion 11

Simulations in Car. Sim • Background and Significance Vehicle Reference Points 4: 1 28 ft-wide v-shaped median • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion Honda Civic (1100 C) crossing at 30 mph with an approach angle of 10 deg 12

Output Data • Background and Significance X XOV XCG • Vehicle Dynamics Analysis Program • Conclusion YCG YS 1 – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results YOV YRP YS Q Y (0, 0) XS 1 XS XRP 13

Graphic Output Minimum of Reference RP 2 Nomograph and Point Maximum 1 2 of RP 1 • Background and Significance – Simulations in Carsim – Output Data – Graphic Output – Main Program Vertical (m) (in) • Vehicle Dynamics Analysis Program • Results • Conclusion Lateral (ft) (m) 14

Graphic Output • Background and Significance • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion 15

Main Program • Background and Significance • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion Hierarchy of the main program investigation of vshaped medians FOR-loops slope 4: 1 6: 1 width 16 ft angle 10 deg speed vehicle 8: 1 . . . 20 ft 15 deg 30 mph 10: 1 12: 1 56 ft 20 deg 40 mph Compact Car 25 deg 50 mph 60 mph Pickup Truck 16

Main Program main_V 2. m • Background and Significance M-functions: ·Clear MATLAB’s workspace ·Launch Car. Sim in an Active. X ·Calculus of speed limit (skipped) ·Initialize the complete log file horizontal road curvature Generate the horizontal geometry of the centerline IF Car. Sim loaded the correct database • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion road. XY_V 2. m EXIT ELSE road. Z_V 2. m THEN Loop FOR slope Generate the elevation of the centerline Loop FOR width IF curve radius is sufficient to accommodate the median width roadd. Z_V 2. m ELSE THEN ·Initialize variables ·Generate the road ·Save the road profile in a txt-file ·Compute the sampling parameters & intervals for this median Loop FOR angle ·Compute the initial position and yaw angle of the vehicle, given the previous conditions of road curvature, median width, and approach angle ·Set up the initial position and yaw angle EXIT Generate the offcenter elevation of the road segments roadcolors_V 2. m Generate the road colors road_V 2. m Set up the main road dataset road 2 txt_V 2. m yawangle_V 2. m 17

Main Program main_V 2. m (cont’d) IF approach angle cannot be reached due to the geometry of the road • Background and Significance THEN SKIP angle Add a new entry to the log file (format txt) for each new error or warning message displayed ELSE Loop FOR speed • Vehicle Dynamics Analysis Program IF speed exceeds limit for this curvature (ignored) – Simulations in Carsim – Output Data – Graphic Output – Main Program logfile_V 2. m THEN SKIP speed ELSE ·Predict the simulation time ·Run Car. Sim for the compact car ·Extract, treat, sample, and save the data ·Run Car. Sim for the pickup truck ·Extract, treat, sample, and save the data ·Compute the min & max of the trajectories ·Generate the envelope ·Generate & save the nomograph(s) cable heights • Results run. Car. Sim_V 2. m savedata_V 2. m savegraph_V 2. m ·Update the complete log file with the contents of the log file created specifically for this median • Conclusion Close Car. Sim® 18

Main Program Road centerline • Background and Significance X Trajectory of the vehicle • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program 1/ p -q 2 a q+a q q • Results • Conclusion Y (0, 0) Y 0 L 0 19

Main Program • Background and Significance • Vehicle Dynamics Analysis Program – Simulations in Carsim – Output Data – Graphic Output – Main Program • Results • Conclusion 20

Results • Background and Significance Outline • Vehicle Dynamics Analysis Program • Overview • Results • Table Headlines – – Overview Table Headlines Nomographs Table Analyses • Nomographs • Table Analyses • Conclusion 21

Overview • Background and Significance • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses • Conclusion 275 nomographs symmetric v-shaped medians 11 widths 5 slopes 5 different curves of radii: • 300 m [984 ft] • 600 m [1, 969 ft] • 900 m [2, 953 ft] • 1, 200 m [3, 937 ft] • 3, 505 m [11, 500 ft] • generic low-tension 3 -Strand Cable barrier • • • 22

Table Headlines • Background and Significance • Curve Radius • Vehicle Dynamics Analysis Program • Slope • Results • Green/Red Zones – – Overview Table Headlines Nomographs Table Analyses • Widths of Green Zones • Number of log entries • Ground plowing • Conclusion 23

Nomographs • Background and Significance • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses • Conclusion 24

Nomographs • Background and Significance • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses • Conclusion 25

Nomographs • Background and Significance • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses • Conclusion 26

Table Analyses • Background and Significance (Sum of) Log File [# of events] Width [ft] Curve Radius [ft] Slope [H: V] 16 20 24 28 32 36 40 44 48 52 56 6 9 13 21 38 46 49 49 54 60 65 410 4: 1 2 5 8 9 9 10 13 13 18 19 20 126 6: 1 1 1 2 8 9 9 9 11 15 83 8: 1 1 2 9 9 9 10 10 70 10: 1 1 1 7 9 9 10 10 67 12: 1 1 1 4 9 9 10 10 64 0 1 2 2 4 7 8 13 26 30 37 130 4: 1 0 1 2 2 4 7 7 9 11 10 11 64 6: 1 0 0 0 1 4 6 9 9 29 8: 1 0 0 0 0 3 5 8 16 10: 1 0 0 0 0 3 3 5 11 12: 1 0 0 0 0 3 3 4 10 0 0 1 2 2 4 3 6 8 9 8 43 4: 1 0 0 1 2 2 4 3 6 7 7 6 38 6: 1 0 0 0 0 1 2 2 5 8: 1 0 0 0 10: 1 0 0 0 12: 1 0 0 0 0 1 1 2 2 4 5 8 27 4: 1 0 0 1 1 2 2 4 5 6 25 6: 1 0 0 0 0 0 2 2 8: 1 0 0 0 10: 1 0 0 0 12: 1 0 0 0 0 1 1 3 3 3 13 4: 1 0 0 1 1 3 3 3 13 6: 1 0 0 0 8: 1 0 0 0 10: 1 0 0 0 12: 1 0 0 0 6 10 17 26 47 60 63 71 95 107 121 623 984 • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses 1969 2953 3937 11500 • Conclusion Total 27

Table Analyses • Background and Significance • Vehicle Dynamics Analysis Program • Results – – Overview Table Headlines Nomographs Table Analyses • The different radii do not dramatically change the cumulative green widths • The most noticeable differences in green width occur for the 300 m radius • The lowest green widths occur in 4: 1 medians of width 32 ft to 44 ft • The most common red/green zones combination is 3 -2 • Conclusion 28

Conclusion • Background and Significance • Database of graphic guidelines for optimum placement of cable barriers • Review of current practices in the US • Vehicle Dynamics Analysis Program • Results • Conclusion • Investigation of the dynamics of run-off-road crossover vehicles • Development of an automated process for modeling and simulation of trajectories across various off-road geometries and conditions • Insights for symmetric v-shaped medians in curves, and analysis of key simulations cases 29

Conclusion • Background and Significance • Vehicle Dynamics Analysis Program • Results • Direct validation with full-scale crash test data • Study of other curvature factors (superelevation, …) • Update with fast evolution of cable barrier systems and ITS • Conclusion 30

Dissertation Director and Academic Advisor: Azim Eskandarian Professor of Engineering and Applied Science Director of the Center for Intelligent Systems Research Advisor: Thank you Dhafer Marzougui Research Director at the National Crash Analysis Center Committee Members: Majid T. Manzari, Professor of Civil Engineering Samer Hamdar, Assistant Professor of Civil and Environmental Engineering Cing-Dao (Steve) Kan, Associate Research Professor