Vehicle Handling with Tire Tread Separation - Transportation Safety

SAE TECHNICALPAPER SERIES 1999-01-0120Vehicle Handling with Tire Tread SeparationCharles P. DickersonCollision Engineering Associates, Inc.Mark W. ArndtTransportation Safety Technologies, Inc.Stephen M. ArndtSafety Engineering and Forensic Analysis, Inc.Reprinted From: Vehicle Dynamics and Simulation 1999(SP-1445)International Congress and ExpositionDetroit, MichiganMarch 1-4, 1999400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760

The appearance of this ISSN code at the bottom of this page indicates SAE’s consent that copies of thepaper may be made for personal or internal use of specific clients. This consent is given on the condition,however, that the copier pay a $7.00 per article copy fee through the Copyright Clearance Center, Inc.Operations Center, 222 Rosewood Drive, Danvers, MA 01923 for copying beyond that permitted by Sections107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying such ascopying for general distribution, for advertising or promotional purposes, for creating new collective works,or for resale.SAE routinely stocks printed papers for a period of three years following date of publication. Direct yourorders to SAE Customer Sales and Satisfaction Department.Quantity reprint rates can be obtained from the Customer Sales and Satisfaction Department.To request permission to reprint a technical paper or permission to use copyrighted SAE publications inother works, contact the SAE Publications Group.All SAE papers, standards, and selectedbooks are abstracted and indexed in theGlobal Mobility DatabaseNo part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior writtenpermission of the publisher.ISSN 0148-7191Copyright 1999 Society of Automotive Engineers, Inc.Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solelyresponsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published inSAE Transactions. For permission to publish this paper in full or in part, contact the SAE Publications Group.Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.Printed in USA

1999-01-0120Vehicle Handling with Tire Tread SeparationCopyright © 1999 Society of Automotive Engineers, Inc.Charles P. DickersonCollision Engineering Associates, Inc.Mark W. ArndtTransportation Safety Technologies, Inc.Stephen M. ArndtSafety Engineering and Forensic Analysis, Inc.ABSTRACTCatastrophic and sudden tire tread separation is an eventthat drivers of motor vehicles may encounter and, insome instances, is implicated as the cause of motor vehiclecrashes and related injury or property damage. In aneffort to understand how tire tread separation affectsvehicle handling, a series of tread separation handlingtest programs were conducted. In each tread separationtest program a sport utility vehicle was instrumented andequipped with steel belted radial tires that were modifiedto emulate tread separation between the inner and outersteel belts. The test vehicle was then subjected to a varietyof open and closed loop handling test maneuvers.This paper presents the data and analysis from thesetests. The research demonstrates through controlledexperiments that a tire tread separation has an effect onthe vehicle’s fundamental handling characteristics. It alsodemonstrates that the effect depends on the position ofthe compromised tire on the vehicle.INTRODUCTIONIn order to better understand the effects of tire tread separationon a vehicle, a series of handling tests has beenperformed on vehicles with tires modified to emulate separationbetween the inner and outer tread belts. A methodologyfor creating a tire with a separated tread that isrepeatable and uniform over multiple tests will be presented.A different methodology for generating a suddenand catastrophic failure was also used and will be presented.Two different test programs were performed. In one, a tirewas modified to emulate complete tread separation andmounted on the test vehicle. A series of steady state andtransient limit performance test maneuvers were thenperformed.In the second series, an actual catastrophic failure wasgenerated. The failures were designed to occur while thevehicle was traveling in a straight line and the subsequentvehicle response was then measured. The tire preparationand test protocols for this series are significantly differentthan those in the first test series. For this reason,the tire failure simulation tests will be described and theresults presented in separate sections.The vehicle used in both test series was a sport utilityvehicle equipped with instrumentation and added safetyequipment to protect the driver. The modifications weremade such that any effects on the mass and inertial propertiesof the test vehicle were minimal.PREVIOUS RESEARCHThe subject of tires and their effect on automobile andtruck performance is a topic that is extensively covered inthe literature. This is expected since for reasonablespeeds, and assuming no collisions with other vehiclesand objects, the significant forces that affect the motion ofthe vehicle are transmitted through the tires. The literatureon tire failure testing is small in comparison to thatconcerned with the effects of tires on performance andhandling of vehicles.In the Experimental Safety Vehicle work of 1974, Jacobsonstated that many high-speed road cross over crashesmight be attributable to tire failures. In the Jacobson testseries a dramatic difference in vehicle behavior betweenfront and rear tire failures was observed. In general, reartire failures were more likely to result in loss of control [1].In 1994, Metz, et. al. presented a series of simulationsconcerned with vehicle evasive maneuver capability withflat tires. The simulations predicted behavior similar tothat presented in this work [2]. Blythe et. al. presented aseries of vehicle handling tests in which a tire failure(blowout) was induced. Their tests indicated that the vehi-1

cle pulls towards the damaged tire and that a rear blowoutresults in a vehicle that exhibits oversteer tendencies[3].Other works describe the importance of tire behavior onvehicle handling and stability. Allen, et. al. state that“Vehicle Handling Stability is dominated by tire forceresponse characteristics.” Allen, et. al. further state thatdifferent front to rear tire saturation effects are the maincause of directional stability problems [4]. In his book,Wong mentions that oversteer is not desirable from adirectional stability point of view [5]. Milliken and Millikendiscuss tire failure briefly stating that a rear tire failurehas a destabilizing effect on the vehicle [6]. In general theliterature consistently states that a vehicle that oversteers,by design or circumstance, is highly undesirable.TEST VEHICLESIn the test programs described, two nearly identical vehicleswere used. Both were two-wheel drive Ford BroncoII’s; one was a 1988 model year while the other was a1989 model year. Prior to configuring the vehicles fortesting, a complete mechanical inspection was performedby certified mechanics. The purpose of the inspectionswas to insure that the drive train, suspension, frame andbody were within original factory specification tolerance.Prior to modifying the vehicles, the center of gravity incurb configuration was measured. This was also performedafter test setup. The results of the “before” and“after” tests are presented in Table 1. Test vehicle setupconsisted of installing instrumentation, data acquisitionand driver protection equipment. In order to maintainvehicle center of gravity and inertial properties, most ofthe interior components were removed during setup.Each component that was removed was weighed and itscenter of gravity measured and recorded relative to thevehicle.Table 1.Test Vehicle Weight, Before and After Setup'88 pre-test '88 test '89 pre-test '89 testfront axle wt. (kg) 784 798 769 831rear axle wt. (kg) 741 749 748 750total (kg) 1525 1547 1517 1581cg height (mm) 678 711 659 678A portable, compact signal conditioning and data acquisitionsystem that was assembled specifically for handlingtesting was installed making the test vehicle fully self contained.Table 2 is a list of the vehicle parametersrecorded. All of the data was captured digitally at a samplerate of 100 Hz with a 10 Hz analog pre-filter.The extra protection added for the driver consisted of afive point harness, a roll bar, window netting and a lightweightoutrigger system. The outrigger design wasreported in a previous paper concerning the effects ofoutrigger design on vehicle handling [6]. Figure 1 showsone of the test vehicles in test configuration.TIRESFor all of the tests Firestone FR-480, size P205/75R15,tires were used. All of the tires used in the test programswere purchased new and throughout the test programstheir condition, as well as that of the rims, was continuallyassessed with tire and/or rim replacements made as necessary.Tire pressure was set to the vehicles’ recommendedvalue and maintained by checking it betweeneach test run. Pressure in the separated tire was set 2 psibelow the recommended value to account for theincrease in volume that occurs when the tread detaches.Prior to beginning a test series at the beginning of theday or after a break, the tires were warmed up by drivingfor approximately twenty minutes. Modified tires were notsubjected to the warmup procedure.Table 2a.Table 2b.Instrumentation Onboard ‘89 Bronco IIChannelParameter1 Velocity2 Heading3 Steer Torque4 Roll Angle5 Steering Wheel Angle6 Left Front Wheel Angle7 Right Front Wheel Angle8 Longitudinal Acceleration9 Lateral Acceleration10 Vertical Acceleration11 Yaw Rate12 Roll RateInstrumentation Onboard ‘89 Bronco IIChannelParameter1 Longitudinal Velocity2 Heading3 Lateral Velocity4 Roll Angle5 Steering Wheel Angle6 Left Front Wheel Angle7 Right Front Wheel Angle8 Longitudinal Acceleration9 Lateral Acceleration10 Vertical Acceleration11 Roll Rate12 Left Front Suspension Position13 Right Front Suspension Position14 Left Rear Suspension Position15 Right Rear Suspension Position2

Through many investigations of tire tread separation failures,it has been observed that typically the tread andouter tread belt separate from the tire and, in manycases, most or all of the tread detaches from the tire. Forthe first test program, tires were modified to emulate thistotal loss of the tread and outer belt.The tires used in these test series have two steel beltsunder the tread. In order to create the modified tire, thetread was first removed using a sharpened putty knifelubricated with soap and water. This was accomplishedby first cutting the base of the tread shoulder laterallyaround the perimeter of both sides of the tire. A cut wasthen made across the tread down to the first steel belt.Using locking pliers it was then possible to grab the looseend of tread piece and peel the tread off of the carcassby pulling and lightly cutting the interface between thetread and the carcass. The outer belt was removed bycarefully cutting between the tread belts and pulling outerbelt wires off of the carcass. At the end of this processthe inner tread belt and carcass remained intact withmost of the rubber between the inner and outer tread beltin place.In addition to the instruments recording vehicle responsedata, video cameras captured vehicle behavior duringeach test run.CIRCLE TURN TEST (UNDERSTEER)The purpose of conducting circle turn tests was to determinethe understeer/oversteer characteristics of the vehicleusing a quasi-steady state maneuver. The maneuverconsisted of driving the vehicle around a 61 m (200 ft.)diameter circle. The test was initiated with the vehicle atrest on a tangent line to the circle and the wheels pointedstraight. The vehicle was then slowly accelerated andsteered to stay on the 200-ft diameter circle. Slow accelerationcontinued until either the vehicle reached maximumspeed due to drive wheel slip or it would no longerstay on the desired path. Typically a test run required 30seconds to complete. Runs were conducted using bothright and left turn maneuvers to determine any differencesin vehicle behavior based on turn direction.STEP STEER TEST (J-TURN)This maneuver was performed by bringing the vehicle tospeed, releasing the throttle, allowing the velocity to stabilizeto the desired value, then rapidly applying a stepsteer input of the desired magnitude. Steer input wasmaintained until a steady state condition was attained oroutrigger contact occurred. The step steer maneuverswere conducted for a single steer magnitude (180degrees) and two speeds (11 and 16 m/s). The purposeof the test was to provide data concerning the transientresponse of the vehicle.OBSTACLE AVOIDANCEFigure 1. Test VehicleFigure 2. Obstacle Avoidance CourseTEST MANEUVERSThe test maneuvers included circle turn (under steer),step steer (J-turn) and obstacle avoidance tests. Theobstacle avoidance tests conducted using the coursedepicted in Figure 2. Braking tests were also conductedto characterize the test surface coefficient of friction. Alltests were run on a flat asphalt or concrete surface.The obstacle avoidance test series was conducted usingthe course shown in Figure 2. The course was entered ata specified speed and the throttle released prior to enteringthe first gate. Throttle and brake were not applied inthe course. Drivers were required to steer through a 2.66m (12 ft) avoidance gate offset 1.33 m (6 ft) from the centerlineof the entry lane, and then steer back to the 2.66m (12 ft) wide exit lane. The avoidance gate was 18.3 m(60 ft) from the end of the entry lane and 18.3 m (60 ft)from the beginning of the exit lane. Cones were used tomark the course lanes and gate. Contact with the conesor failure to stay in the lanes constituted a failure.Multiple attempts to drive the course were made at varyingspeeds to increase the possibility of completion anddocument the required driver input for completion. Testswere conducted for two vehicle tire configurations: fourunmodified tires and modified tire on the right rear. Runswere made at gradually increasing speeds until consistentfailure to negotiate the course was observed. The leftor right label refers to the initial turn direction of the test.3

DISCUSSION - CIRCLE TURN TEST RESULTSThe circle turn maneuver was performed with fourunmodified tires, 3 unmodified tires with a modified tireon the left rear and 3 unmodified tires with a modified tireon the left front. Turns were made both towards (left) andaway (right) from the side with the modified tire.FOUR UNMODIFIED TIRESThe Bronco II equipped with four unmodified tires understeeredthroughout the circle turn test. In the limit, thefront axle saturated prior to the rear and the vehicle“plowed out.” These observations were noted for bothright and left circle turns. The maximum lateral accelerationachieved was approximately 0.74 g and 0.78 g for theright and left circle turns respectively. As can be seen inthe Figures 3 and 4, the Bronco II understeer gradientincreased as the maximum lateral acceleration wasachieved.MODIFIED TIRE AT LEFT REAR POSITIONFigure 3. Left Circle Turn ResultsAnalysis of the circle turn data was performed using thesteering wheel angle and lateral acceleration data to plotthe oversteer/understeer characteristics for the vehicle ineach test configuration.When a modified tire was placed on the left rear wheelposition and the circle turn was conducted to the right,the difference in behavior was dramatic. Figure 4 showsthat the vehicle understeered slightly until a lateral accelerationof approximately 0.3 G was attained. The vehiclethen began to oversteer at an increasing rate. The maximumlateral acceleration achieved during this test wasapproximately 0.6 G.When the test was run to the left, the vehicle behavedsimilarly to the vehicle with four unmodified tires. Thevehicle did not understeer as much, but was able toachieve a maximum lateral acceleration of 0.74 G whichwas only slightly less than the 0.78 G achieved with fourunmodified tires. Vehicle understeer increased as the lateralacceleration increased. The test was terminatedwhen the vehicle velocity could not be increased due toinside rear wheel spin. Neither vehicle was equipped witha limited slip type rear differential.MODIFIED TIRE AT LEFT FRONT POSITIONFigure 4. Right Circle Turn ResultsThe steering wheel angle data was first normalized bydividing by the steering box gear ratio of the Bronco II.Both data sets were then filtered using a 100 point movingaverage. The combined data set (steer angle v. lateralacceleration) was then curve fitted with a fifth order polynomialand plotted in Figures 3 and 4.A circle turn test to the right with the modified tire on theleft front wheel position resulted in the Bronco II understeeringsubstantially more when compared to the fourunmodified tire configuration. The maximum lateral accelerationachieved during this test was approximately 0.55G. This was a reduction in the lateral acceleration ofapproximately 0.19 G. The test was terminated when thefront end of the vehicle began to plow out.When the test was repeated to the left, the vehiclebehaved similarly to the vehicle with four unmodifiedtires. The understeer was slightly greater than with fourunmodified tires and the maximum lateral accelerationachieved was slightly less at approximately 0.72 G ascompared to 0.78 G. For the Bronco II at the lateralacceleration limit, the inboard front wheel typically liftedoff of the ground thus effectively eliminating the contributionof the modified tire. This test was terminated whenthe vehicle could no longer be driven on the circle due tofront axle saturation4

DISCUSSION - STEP STEER TEST RESULTSStep steer test runs were made with a steer (at steeringwheel) input of 180 degrees and at speeds of 11 m/s (25mph) and 16 m/s (35 mph). Runs were made in both leftand right directions with three vehicle tire configurations:four unmodified tires, modified tire on the left front andmodified tire on the left rear. Portions of the resulting dataare presented in Figures 5 through 12. The data hasbeen smoothed by running it through a nine sample widemoving average window. Each plot presents a singlemaneuver with data from each of the vehicle configurationsoverlaid for comparison.RIGHT STEP STEERAs expected the right turn maneuvers resulted in themost dramatic difference between the three configurations.In the right steer series the steer input was awayfrom the modified tire resulting in lateral load transferonto the modified tire. The steer input and longitudinalvelocity plots demonstrated that the three runs presentedoccurred at approximately the same initial speed and thatthe steer inputs were nearly identical. Once the maneuverbegan, the longitudinal velocity of the modified reartire configuration dropped off rapidly due to the resultinghigh slip angle. At 11 m/s the unmodified tires and frontmodified configurations behaved similarly. At 16 m/s theunmodified tires configuration slowed more rapidly thanthe front modified configuration.The difference between the tire configurations was bestdemonstrated in the lateral velocity and slip angle plots.The modified rear tire configuration spun out at bothspeeds. The discontinuity in the plot is due to the vehicleexceeding the maximum slip angle allowed by the sensor(approximately 40 degrees). At 16 m/s the slip angle ofthe four unmodified tires configuration was more thantwice that of the modified front tire configuration while at11 m/s the slip angle response was quite similar. At bothspeeds the four unmodified tires and front modified configurationsdemonstrated stable understeer responses tothe steer input. At both speeds the modified rear tire configurationdemonstrated an unstable response.The heading and yaw rate data followed the same trendwith the yaw rate, and naturally, the heading changebeing significantly greater for the rear modified condition.At both speeds the heading change and yaw rate weregreater for the four unmodified tires configuration than forthe modified front tire configuration indicating greaterundersteer for the modified front tire configuration.The lateral acceleration results showed the four unmodifiedtires configuration to generate higher values than themodified front tire configuration. The modified rear tireconfiguration showed a higher value than the fourunmodified tires configuration in the 11 m/s run andapproximately the same value in the 16 m/s run. Theanomalies observed in the results for the modified reartire configuration data were caused by the high yaw rategenerated in the maneuver.LEFT STEP STEERComparison of the left to right data was more interestingthan comparison of the different configurations in the leftstep steer runs. Again the steer input and longitudinalvelocity data shows the initial conditions of the individualruns to be approximately the same. The first significantobservation was that all of the configurations demonstrateda stable understeer response at both speeds.At 11 m/s all three configurations had similar responses.The four unmodified tires and modified rear tire configurationswere almost identical for the parameters measured.The modified left front configuration displayedlower heading change, yaw rate and lateral accelerationresponses than the other two configurations.At 16 m/s the modified rear tire configuration differs fromthe four unmodified tires and modified front tire configurations.The lateral velocity and slip angle responses weregreater. The heading change and yaw rate were similar tothose of the four unmodified tires configuration. The modifiedfront tire configuration displayed less headingchange.RESULTS - OBSTACLE AVOIDANCE TESTDue to the oscillatory nature of the steer input required tonegotiate the test course, vehicle response was similarlyoscillatory in nature. In general an initial three-peak curveof steering input was necessary to complete the obstacleavoidance course. Overshoot during the recovery afterentering the return lane resulted in additional steeringpeaks. Most of the measured vehicle parameters had asimilar oscillatory nature that differed between the differenttire configurations. By measuring the magnitude ofthe vehicle response peaks, a family of graphs has beenproduced. Graphs for the steer input versus initial speedand the vehicle responses of yaw angle and yaw anglerate versus initial speed are shown in Figures 13 through15.The graphs have points labeled as "modified tires outsideor inside." The use of the term outside or inside refers tothe location of the modified rear tire in the first turn of themaneuver. A test with the "modified tire outside" refers toa course with the obstacle to the left.While the early response peaks were similar for all testconditions, a failure to negotiate the obstacle avoidancecourse often resulted in a higher magnitude of vehicleresponses. In general the population of vehicleresponses was different depending on the rear tire condition.5

DISCUSSION - OBSTACLE AVOIDANCE TESTWhile obstacle avoidance tests are by nature subjectivethey also yield objective data. The test method dependson the driver’s input. The simple pass/fail criteria whichhas been applied to a vehicle as a result of obstacleavoidance testing is not appropriate given the sensitivityof the test method to driver skill and driver learning duringtesting.Pass/fail in the negotiation of the obstacle avoidancecourse is an objective indicator for comparing test resultswhen measuring driver control and vehicle response. Theobstacle avoidance course could be completed at higherspeed with unmodified tires. Without a modified tire, thecourse could not be successfully negotiated above 16 m/s (35 mph). Successful completion of the obstacle avoidancecourse could not be accomplished for the conditionof a modified tire on the inside rear above 13.4 m/s (30mph).Since driver input improved as an obvious result ofrepeated attempts at the obstacle avoidance course, apoint of comparison was derived in comparing only thesuccessful runs. Overall, the data indicated that a moredeliberate input for the first two steer inputs and a greatermagnitude for the third steer input were necessary tocomplete the obstacle avoidance course when the modifiedtire configuration was compared to the unmodifiedtires configuration. This result is illustrated in the family ofthree graphs in Figure 13. The higher third peak steermagnitude for the modified inside tire test was consistentwith the driver’s expectation of vehicle oversteer duringthe second turning maneuver.The vehicle yaw angle behavior is shown in Figure 14.For the first two peaks, the modified tire vehicle yawangle was higher in magnitude that for the vehicle withfour unmodified tires. The first peak graph demonstratesthat for a “passing” run with the modified tire on theinside, the vehicle was within the population of responsefor unmodified tires. If the modified tire was on the outsideof the initial turn, a significantly higher yaw anglewas observed. This result was consistent with the unstableoversteer condition which existed with the modifiedrear tire on the outside of a turning maneuver.The family of graphs showing peak yaw rate responsemore clearly demonstrates the generally higher magnitudeof vehicle response that occurs with the modified tireon the rear. Similar to yaw angle, the first peak yaw rateresponse plot shows the maneuver with the modifiedinside tire within the population of unmodified tire results.In general the higher magnitude of yaw rate peakresponse is more pronounced than that observed for yawangle.Driver response to a vehicle’s behavior during obstacleavoidance testing provides an opportunity to compare thenecessary driver input for successful completion of thetest course. Failed test results show the greatly variedresponse of the vehicle and driver. The oversteer conditionwhich exists when modified tires are attached at therear is objectively demonstrated in the plots shown in figures13 through 15. Overall different driver response wasnecessary in the presence of the modified tire. Differentsteer input magnitudes and greater vehicle responseswere necessary to maintain control of the vehicle.TIRE FAILURE SIMULATIONTEST VEHICLE – In this test series, the 1989 two-wheeldrive Ford Bronco II was used. It was prepared in exactlythe same way as the test series described above. In additionto the instruments already onboard, a video camerawas mounted internally with two microphones to capturethe sounds of the tire failure. A pink noise generator andaudio noise power level meter were used to calibrate thesound recording. The steering shaft between the handwheel and the steering box was also instrumented tomeasure torque.TIRE PREPARATION – The tires were prepared somewhatsimilarly to the total separation tires with the exceptionthat the tread was not removed. The tire wascarefully cut in from the edge of the base of the treadshoulder between the two outer steel belts. This cut wentaround the entire perimeter of the tire but not all of theway across the tread block. Cuts were made goinginward from both tread shoulders leaving a narrow patchof rubber at the center of the tread block bonding theinner belt to the outer belt around the perimeter of thetire. The tread was also scored parallel to the outer beltfilaments over its entire width in one place. It should benoted that this process evolved over several test runs aschanges were made to create a tread separation failurein a predictable time and place.TEST PROTOCOL – The modified tire was mounted onthe right rear wheel position. The test was run on the taxiwayof a local airport. This provided a long area for accelerationand adequate lateral space for vehicle motion.The test driver was instructed to accelerate to the desiredtest speed and maintain as room allowed. When thetread separation occurred, the test driver held the steeringwheel constant and did not respond until it was necessaryso as to avoid leaving the taxiway.TEST RESULTS – Three successful runs were made.Success was defined as a run in which the tread separationwas sudden and the majority, if not all, of the treaddetached and electronic data was captured. Data fromthese runs are presented in Figures 16 through 18.Speed, steering wheel angle, and longitudinal and lateralacceleration are presented.Overall, the separation resulted in a loud banging noiseas the loose tread battered the fender and inner wheelwell. Accompanying the noise was damage to the rightrear quarter panel and tailpipe. The driver reported thatthe noise was very loud. In all cases the tire carcassremained inflated after the separation was complete. In6

Figure 5. 11 m/s Left Step Steer Data9

Figure 6. 11 m/s Step Steer Data10

Figure 7. 11 m/s Right Step Steer Data11

Figure 8. 11 m/s Right Step Steer Data12

Figure 9. 16 m/s Left Step Steer Data13

Figure 10. 16 m/s Left Step Steer Data14

Figure 11. 16 m/s Right Step Steer Data15

Figure 12. 16 m/s Left Step Steer Data16

Figure 13. Obstacle Avoidance Steering Response Peaks17

Figure 14. Obstacle Avoidance Yaw Angle Response Peaks18

Figure 15. Obstacle Avoidance Yaw Rate Response Peaks19

Figure 16. Time Failure Simulation, Run A20

Figure 17. Time Failure Simulation, Run B21

Figure 18. Time Failure Simulation, Run C22