Newsletter Volume 4 - 4th Quarter 2004

Human Motorcycle Control: Stabilizing and De-Stabilizing, Part I

Case Study: Human Motorcycle Control: Stabilizing and De-Stabilizing, Part I

Software: LifeMOD™ v2005 New Product Release!

Other News: New Publication: "Evaluation of a Computer-Simulation Model for Human Ambulation on Stilts"

In this issue we will examine a motorcycle crash from the standpoint of the rider's actions which led to the unsafe handling of the motorcycle. In the next issue of the newsletter we will examine the biomechanics of the resulting motorcycle crash and study various injury mechanisms.

Learn how BRG.LifeMOD™ makes this technology accessible to all researchers and students of biomechanics. Additionally, we are pleased to announce the release of LifeMOD™ version 2005 with many new functions and features. We invite you to download a free trial version of the software; try out one of our 17 easy-to-follow tutorials and begin building physics-based human models today.

HUMAN MOTORCYCLE CONTROL:
STABILIZING AND DE-STABILIZING, PART I

Introduction

A motorcycle rider must be constantly attentive to environmental disturbances - potholes, insects, wind, etc. These disturbances can cause crashes, and crashes usually result in injury.

The next two issues of this newsletter will explore the event of an injury-producing motorcycle crash, both from the standpoint of interpreting the rider's actions to avoid a road obstacle during a lane change maneuver at 50 mph, and the evaluation of the biomechanics response of the rider to the resulting crash conditions [McGuan, 93].

In this issue of the newsletter, we will examine the actions of the rider which led to the unsafe handling of the motorcycle.

Rider-Cycle Mechanical System

Manual control of motorcycles has long been of practical and theoretical interest. Since motorcycles or any 2-wheeled vehicles can be susceptible to environmental disturbances and require constant rider attention, they present unique problems in stability and control.

To analyze the actions of the rider for motorcycle stabilization, the rider-cycle system must be viewed as a mechanical system, with the motorcycle as the controlled element, and the rider as the controlling element.

A 2-wheeler in its simplest form is a free-roaming vehicle with only one degree-of-freedom under the direct control of the rider. That degree of freedom is roll, and it is apparent by inspection that if the dynamics of the system are neglected, the relationship between vehicle speed (V), roll angle (R) and radius of curvature of the vehicle path (C) is:

Eq. 1

The rider acts as a roll controller, evaluating the required roll angle for the selected path and submitting roll acceleration demands to the system, mainly in the form of steering torque. Body lean angle is an ineffective form of primary control input; it has a low gain, and large body lean angles disturb the machine significantly if they are applied rapidly. Skilled and experienced riders brace themselves against the machine using their outside leg to support rapid roll rates and minimize the interference from upper body dynamics with the machine behavior [Rice, 78] as can be observed in the animations on the right side panel.

This trivial representation of the motorcycle is not useful for any reasonable type of scientific inquiry. The dynamics of the system cannot be ignored and the active intervention of the rider is necessary not only for control but also for stabilization. For a reasonable inquiry, the modeling approach will involve a human model coupled to a full vehicle model of the motorcycle, complete with tire forces. A closed loop controller will be used to join the actions of the controlling element (rider) with the controlled element (motorcycle).

Motorcycle Model

The vehicle dynamics can be characterized analytically via a system of differential equations and studied using classical mechanics. Practical limitations necessitate the usage of "automated" mechanics programs which generate and solve the equations of controlled motion. For this inquiry, the motorcycle will be modeled using MSC/ADAMS (MSC software).

The rider-cycle system can be viewed as a closed-loop mechanical system. The controlled element, the motorcycle, consists of four parts: the two wheels, the steering assembly, and the body, connected with revolute joints. The effects of the suspension system are not included in this inquiry. This results in a system with eight degrees-of-freedom, including the six components of vehicle gross motion and the rotational freedoms of the front wheel and the steering assembly.

This representation yields three complex eigenvalues:

Wobble mode, (flutter mode) involving a response of the front fork assembly about its hinge point. This mode is always present even under rider control.
Weave mode, involving the coupled roll and yaw motions of the motorcycle. This mode is always present even under rider control.
Capsize mode, relating to vehicle roll. This mode may gradually diverge in the absence of rider control.

The rear wheel is driven with a torque driver resulting in a vehicle speed of 50 mph. The inertia of the rotating masses of the wheels helps to dampen the roll mode of the motorcycle. The forces resulting from the tire/road interaction are generated using a tire force algorithm which generates the non-linear 6 force/torque components at the tire contact patch. The non-linear force transmission between the tire and road introduce a response delay, or a phase lag effect the eigenvalues in interesting ways. [Sharp, 71]

Human Model

The human model was built using BRG.LifeMOD™ [McGuan, 04]. The segment geometry, mass properties and joints were all generated for a 50 percentile male from internal anthropometric databases. For a human model that will both control the motorcycle and respond to the crash, the joints of the model consisted of both active elements (controlling phase) and passive elements (reactive phase). See figure 1.

Figure 1: Human joints switch from active controlling to passive response, depending on the state of attachment of the rider to the motorcycle.

The active elements provide the rider-initiated steer torque and body lean to maintain the desired path heading and the stabilize the vehicle and the passive elements consist of individual joint limits, stiffness/damping effects and friction/hysteresis derived from the Hybrid III crash dummy. The active and passive modes of rider model are switchable during the simulation, depending on the status of attachment between the rider and the motorcycle model.

The human model was then posed and positioned on the motorcycle model. The interface between the human model and the motorcycle was modeled using contact elements between the rider's feet and foot pegs, pelvis and seat, inner thighs and the vehicle frame. Break-away attachments (springs) are used between the hands and the grips and will release at a certain threshold [Mathiowetz, 86] to allow the human model to detach from the motorcycle during a crash event.

Active Control and Stabilization

For a statically unstable system to be controlled to follow a path, a dynamics compensator feedback loop representing the rider's actions is necessary to sense the output from the motorcycle including heading (H), path lateral variance (D), and roll angle (R), and to provide inputs to the motorcycle including torque steer (T) and rider lean angle (L). Since the capsize mode is mildly divergent, continuous rider control is necessary to produce the desired handling performance.

Figure 2: Closed-loop controller representing the rider-cycle system.

The dynamic compensator chosen for this inquiry is the multiple loop, parallel structure type presented in [Weir, 78] and displayed in figure 2. It functions with an inner loop which serves to stabilize the roll of the vehicle (R) by coupling the states of motorcycle roll angle to rider-initiated steer torque (T). The outer loop serves to maintain the desired heading by coupling the states of motorcycle heading (H) and path variance (D) to rider lean angle (L).

A first order filter is applied to the motorcycle roll angle to produce a time delay in rider response inherent in human neuromuscular actuation. Equation 2 phase shifts the roll angle (R) through the use of a time lag constant (τ). For a healthy, alert rider the time delay (τ) is about 0.1-0.3 seconds [Eaton, 73].

Eq. 2

Methods to tune the gains for the controllers are detailed in [McGuan, 93].

Case 1: Stabilized Maneuver Without Disturbance

To create a baseline simulation, the lane change maneuver is performed for the steady state condition without a disturbance.

To examine the rider/cycle system behavior, the time histories of the control input and the systems responses are studied in figure 3.

Figure 3: Steady-state lane change maneuver.

Phase 1: The maneuver is initiated by the open-loop application of a steering torque to the right, causing small out-tracking steer angle to the right to develop.
Phase 2: With this forced torque input, the motorcycle begins to roll to the left and yaw to the left.
Phase 3: Within the 0.3 second time delay of sensing the path deviation due to the out-tracking, the rider begins to lean to the right, this causes the motorcycle to accelerate the roll rate to the left and the steer angle also goes to the left. With this shift in steer angle, the compensating steer torque also changes sign.
Phase 4: Peak cornering to the left occurs (roll angle and yaw rate are at maximum values).
Phase 5: The path error becomes large as the right turn phase occurs, causing the rider to lean to the left.
Phase 6: This causes the motorcycle to roll to the right and change sign in the steer angle. A large amount of steer torque is used to compensate for the lean to the left during this right cornering portion.
Phase 7: The maneuver is completed, with some path overshoot. The steer torque and the rider lean inputs approach zero.

Case 2: Stabilized Maneuver With Disturbance

In the second case, a disturbance is introduced by simulating the front tire going over a pothole (50 mm) at the apex of one of the turns in the maneuver. This represents a lateral disturbance to the rider-cycle model, and the control system must stabilize the roll motion to prevent it from capsizing.

Figure 4: Stabilized maneuver with pothole disturbance.

The pothole occurs at the point when the motorcycle is at maximum roll, which causes an out-of-plane force to the front tire, and subsequently a lateral disturbance to the rider/cycle system. The effects of the control input (steer torque) and the motion response (motorcycle roll) for the stabilizing action are displayed in figure 4.

In this figure, the disturbance to the roll motion is evident. The torque curve displays the corrective action taken by the compensator with the excited wobble mode superimposed. The 0.3 second time lag inherent in the rider neuromuscular system can be observed by comparing the roll curve to the torque curve. Although the wobble mode is not damped, the roll disturbance is stabilized in a well damped manner.

Case 3: De-Stabilized Maneuver With Disturbance

In the next issue of this newsletter, we will discuss the possibility of rider inattentiveness as a source of error resulting in instability and an injury-producing crash.

References

[]Eaton , D.J., 1973, "Man-Machine Dynamics in the Stabilization of Single-Track Vehicles." University of Michigan, Ph.D. Thesis.
[]Mathiowetz V., et al., 1986, "Grip and pinch strength: norms for 6- to 19-year-olds." Am J Occup Ther. 1986 Oct;40(10):705-11.
[]McGuan, S. 2004 BRG.LifeMOD™ 2004 Users Manual, Biomechanics Research Group, Inc.
[]McGuan, S.P., 1993, "Active Human Surrogate Control of a Motorcycle - Stabilizing and De-Stabilizing" Journal of Passenger Cars.
[]Rice, R.S., 1978, "Rider Skill Influences on Motorcycle Maneuvering." Paper 780312, SAE Congress and Exposition, Cobo Hall, Detroit, MI.
[]Sharp, R.S., 1971, "The Stability and Control of Motorcycles." Journal of Mechanical Engineering Science, Vol. 13, No. 5.
[]Weir, D.H., Zellner, J.W., 1978, "Lateral- Directional Motorcycle Dynamics and Rider Control,” Paper 780304, SAE Congress and Exposition, Cobo Hall, Detroit, Michigan.

SOFTWARE

The BRG is pleased to announce the release of LifeMOD™ version 2005. This new release makes state-of-the-art human modeling accessible to every investigator interested in the physics behind human motion. The release includes a complete code optimization to reduce modeling time by a factor of 10. It also includes many new features and functionality to run many simulation trials from motion data sets. This is very useful for ergonomics studies.

Due to the tremendous response we received from our users in our "learning" muscles, we have introduced several more muscle groups to ensure LifeMOD's position as the most versatile full-body human modeling package available today.

This new version is a direct result from an ongoing and rigorous user dialogue, partnerships with our research community, and the inclusion of much functionality developed by our professional staff to solve the world's most demanding biomechanics issues. View Examples.

SERVICES

The Biomechanics Research Group Inc. is a service-based organization chartered to empower our customers to capture a level of ROI from their technology investment in ways they've never imagined. We are committed to customer service, product excellence and continuous quality improvement in all we do. We provide training, modeling and simulation expertise. Contact us for more information.

OTHER NEWS

New publication: "Evaluation of a Computer-Simulation Model for Human Ambulation on Stilts", C.S. Pan, et al., Journ. Mechanics in Medicine and Biology. Sept., 2004 See PDF

Check out our new model repository! We would like to sincerely thank those who have contributed to the LifeMOD body of knowledge. We pledge to do our best to increase the technical capabilities of LifeMOD™ while developing new ways to educate the community.

Shawn McGuan (BRG CEO) will address the 2005 Korean Cad/Cam Conference as the keynote speaker, Jan 27-29. More information.

If you need further information on our software and services, please contact us.

Primary links