This paper discusses the development of a driver model designed to mimic human behavior for any kind of driving situation. A new concept is introduced to model human short term control. It is based on a strict separation of geometric dynamic planning and plan to action mapping. Such a division enables one to model the effects motivation and knowledge independently. Moreover, it facilitates the task of debugging since errors resulting from faulty geometric dynamic planning can be isolated from those that result from lack of knowledge about the system being controlled. This article is thought to give an overview of a driver system that is novel in its structure. This approach establishes vehicle control in a way that is different from traditional linear systems. A method is introduced that allows one to model multiple aspects of motivation and combine them in a sound manner for the task of creating a geometric dynamic plan. Motivations are divided into two classes: target motivations and limit motivations. These kinds of motivation are represented by a method to construct target and limit maps . A discussion about how to combine these maps and find the optimum geometric dynamic plan concludes the discussion about geometric dynamic planning. Plan-to-action mapping designates the process used to find the appropriate physical values by which the driver controls the vehicle, i.e. the front wheel angle velocity and the longitudinal tracking force. This is the point where knowledge about the system must be included. Simulation results show how to distinguish between errors arising from geometric dynamic planning and errors arising from plan-to-action mapping. Finally, results demonstrate the ability of the driver to maintain control over the vehicle system even at the limit of adhesion.

This paper investigates robust control problem of structural vibrations using shape memory alloy (SMA) wires as actuators. The mathematical model for these SMA actuators is derived with emphasis in model uncertainty. The linearization of the relation between stress and temperature dynamics of SMA actuators is analyzed for active control. To handle the uncertainties caused by the linearization and the neglected high frequency dynamics, optimal H ∞ control was employed to design a controller. An example is used to demonstrate the design procedures and the control system is tested in a nonlinear environment.

In many applications, digital valves driven from stepping motors are often characterized by quantitative errors and in some cases, slow response. A new means of direct digital control is introduced for a spool valve actuated by a stepper motor. With this control strategy, both excellent speed of response and accuracy are simultaneously sustained for the valve. By way of illustration, the characteristics of a digital spool valve are theoretically and experimentally investigated. This paper also deals with the design of the controller and some concepts concerning the digital control of a valve, such as initialization, false protection, etc. An example is given to demonstrate the effectiveness of this digital control approach for a practical electro-hydraulic control system.

This paper presents an approach to vehicle control based upon the paradigm of artificial potential fields. Using this method, the dynamics of the vehicle are coupled with the environment in a manner that ensures that the system exhibits safe motion in the absence of driver inputs. The driver remains in control of the vehicle, however, with the control systems presenting a predictable and safe set of dynamics. With the control approach presented here, integration of various assistance systems can be easily achieved through simple superposition of individual potential and damping functions. A simple example of a combined lanekeeping and stability system demonstrates how this can be accomplished. Preliminary simulation results suggest that both safety and driveability are achievable with such a system, prompting further investigation.

This paper addresses the problem of detecting multiple faults for the longitudinal control system of an automated vehicle. An existing fault diagnostic system which can isolate all single faults is extended to the diagnosis of multiple faults via improved residual processing in the form of fuzzy logic. The new diagnostic system is shown to correctly detect and isolate all single and multiple faults in a subset of the automated vehicle control system components.

Utility spark ignition gasoline engines are deployed worldwide for a variety of functions including transportation, power generation, and fluid movement. These engines are typically air-cooled and employ cost-effective carburetors and fixed angle spark ignition systems. Although utility engines are generally operated on the same gasolines available for highway vehicles, there are remote applications where gasoline formulations may be substantially degraded. Until recently, two and four stroke utility engines have not been a focal point for performance enhancement or pollution control technology. However, the growing awareness of small engine pollution, as well as legislated federal and state requirements, presents many engineering challenges. In this paper, an analytical and empirical model is presented to describe the behavior of internal combustion spark ignition utility engines. The nonlinear engine model will support the design of model-based and intelligent control algorithms prior to dynamometer testing. An electronic control system architecture is introduced using a multi-purpose programmable engine controller which is capable of regulating fuel injection and electronic spark ignition subsystems.

A new approach to chatter control in milling is presented in this paper. The proof-of-concept control system comprises a tool holder, analog and digital control electronics, and power amplifiers to drive the actuator elements. The active tool holder, designed to impart counter-vibration forces to the milling tool, mounts to existing machines with a standard industrial interface. Sensors and piezoelectric actuators are imbedded in the stationary portion of the tool holder and are therefore fixed relative to the body of the milling machine. The controller operates on the two sensor signals, producing two orthogonal actuator drive signals to oppose resonant tool vibrations induced from the cutting forces. The paper first introduces the fundamental concepts of milling chatter and their relation to the described active system. The actuation, sensing and controller details follow. The influence of the tool holder on system dynamics and cutting stability is also addressed. Cutting test results using a titanium alloy are then described, demonstrating an improvement of a factor of five in surface finish relative to the uncontrolled, chatter-dominated case.

There are many different networks which have been promoted for use in manufacturing control systems. In this paper, we discuss the features of three candidates: Ethernet (CSMA/CD), ControlNet (Token Bus) and DeviceNet (CAN Bus). We consider how each of these control networks could be used as a communication backbone for a networked control system. A detailed discussion of the medium access control protocol for each network is provided. The medium access control protocol is responsible for providing both the satisfaction of the time-critical/real-time response requirement over the network and the quality and reliability of the communication between devices on the network. For each protocol, we derive the key parameters of the corresponding network when used in a control situation, including bandwidth, magnitude of the expected time delay, and variation in time delays. Simulation results are presented for several different scenarios, and the advantages and disadvantages of each network are summarized.

We have used observations of posture and movements in biological limbs to derive a controller for an artificial mechanism. The controller architecture emulates some of the known relations between spinal cord circuitry and the musculoskeletal system of vertebrates and, specifically, of the rat. This work relates to recent experiments suggesting that the neural circuitry of the spinal cord may be partitioned into a small set of functional modules. Activation of these modules, each connected to a set of limb muscles, resulted in force fields that have been measured at the endpoint of a limb. These force fields map each position of the foot into a corresponding static force vector. The force fields have been found to converge toward equilibrium positions located inside the leg’s workspace. The experimental observation that vector fields induced by multiple stimulations add vectorially, suggested that convergent force fields form a system of building blocks (or “primitives”) for the generation of stable postures and movements. To emulate this biological mechanism in the control of an artificial two-joint limb, we established relationships among three hierarchical levels — spinal modules, muscles, and actuators — by deriving the mappings among the respective output fields. These mappings are used in combination with an inverse model of the actuators to calculate the actuator commands that generate a desired force field. We tested the ability of this control system to reproduce the force fields generated by the leg muscles of the rat and a set of force fields with significant geometrical features. Our results show that we can successfully and reliably transfer to our artificial system the features of muscle force fields. In addition, we exploited the same principle of vector summation observed in the biological system to combine these muscle fields into a variety of force field patterns, including the gradients of Gaussian potentials and locally parallel fields. We consider this a first step in the generation of a biomorphic motor control system. This work is supported by ONR grant N00014-95-1-0571 and NIH grant MH48185.

In analogy to the flow of fluids, it is expected that the aggregate density and the velocity of vehicles in a section of a freeway adequately describe the traffic flow dynamics. The conservation of mass equation together with the aggregation of the vehicle following dynamics of controlled vehicles describes the evolution of the traffic density and the aggregate speed of a traffic flow. There are two kinds of stability associated with traffic flow problems — string stability (or car-following stability) and traffic flow stability. We make a clear distinction between traffic flow stability and string stability, and such a distinction has not been recognized in the literature, thus far. String stability is stability with respect to intervehicular spacing; intuitively, it ensures the knowledge of the position and velocity of every vehicle in the traffic, within reasonable bounds of error, from the knowledge of the position and velocity of a vehicle in the traffic. String stability is analyzed without adding vehicles to or removing vehicles from the traffic. On the other hand, traffic flow stability deals with the evolution of traffic velocity and density in response to the addition and/or removal of vehicles from the flow. Traffic flow stability can be guaranteed only if the velocity and density solutions of the coupled set of equations is stable, i.e., only if stability with respect to automatic vehicle following and stability with respect to density evolution is guaranteed. Therefore, the flow stability and critical capacity of any section of a highway is dependent not only on the vehicle following control laws and the information used in their synthesis, but also on the spacing policy employed by the control system. Such a dependence has practical consequences in the choice of a spacing policy for adaptive cruise control laws and on the stability of the traffic flow consisting of vehicles equipped with adaptive cruise control features on the existing and future highways. This critical dependence is the subject of investigation in this paper. This problem is analyzed in two steps: The first step is to understand the effect of spacing policy employed by the Intelligent Cruise Control (ICC) systems on traffic flow stability. The second step is to understand how the dynamics of ICC system affects traffic flow stability. Using such an analysis, it is shown that cruise control systems that employ a constant time headway policy lead to unacceptable characteristics for the traffic flows.

This paper presents the design and development of a Multi-module Deployable Manipulator System (MDMS) as well as a dynamical formulation for it. The system is designed for experimental investigations aimed at dynamics and control of this variable geometry manipulator by implementing different control algorithms to regulate its performance. The manipulator operates in a horizontal plane and is unique in that it comprises of four modules, each of which has one revolute joint and one prismatic joint, connected in a chain topology. Each module has a slewing link of approximately 20cm length and is capable of extending by 15cm. The manipulator design involves the selection and sizing of actuators, the design of mounting and connecting components, and the selection of hardware as well as software for real-time control. The dynamical model is formulated using an O ( N ) algorithm, based on the Lagrangian approach and velocity transformations. The O ( N ) character is computationally efficient permitting real-time control of the system.

The international competition demands a continuous improvement of the process in quality and productivity and at the same time a reduction of costs of the product. An important step to reach this target is to increase the degree of automation. Changing into highly automated and simple-operated production processes, requires a secure process-controlling. With a reproducible use of measuring devices the certainty of a high workpiece quality lasts. For realising this reliable process controlling, a qualified system for process monitoring which uses adapted sensors and control strategies is necessary. To acquire automatically all the quality parameters relevant for ground parts as exact and fast as possible and non-destructive as well, a so-called flexible measurement cell was developed. All parameters measured by the flexible measurement cell are used to build up a control system for the grinding process of rotational symmetrical workpieces. For the first time the possibility to control a grinding process by using the subsurface characteristic as a command variable is offered. This system is independent of the grinding wheel specification. Tests have proven the efficiency by the application of CBN as well as corundum. The demand for a quality control and documentation beside the development of tough processes will be very important to process monitoring. For this reason a prototype system acting with an automatic control has been developed.

The Graduate Automotive Technology Education (GATE) Center at the University of Tennessee, Knoxville (UTK) offers courses addressing the simulation, modeling, and control system design of hybrid electric vehicles (HEV). In the Spring of 1999 such a course was conducted to support the UTK FutureCar Challenge entry for 1999. The vehicle modeled is a Dual-configuration Hybrid Electric Vehicle (DHEV) which uses a planetary power-split device similar to the Toyota Hybrid System used in the Toyota “Prius”. The goals of the course included the development of a real-time simulator that could incorporate actual vehicle control hardware in the simulator loop. This “control-hardware-in-the-loop” (CHIL) configuration was used for simulation, control system design, and troubleshooting. This approach allows the simulation of normal vehicle operating conditions as well as emergency fault handling situations in which it may not be desirable to subject the actual prototype vehicle to a given test condition. Additionally, it is possible to do a great deal of control system testing and development without an operating vehicle.

Loop heat pipes currently are being used in the thermal control systems for satellites. To expand possible loop heat pipe applications, information regarding response to transient heat inputs is required. In this investigation, two loop heat pipes with dual compensation chambers were subjected to heat inputs of varying magnitude, frequency, and waveform (square and sinusoidal). The performance of each loop heat pipe under these conditions was evaluated in different gravitational orientations. The upper and lower limits of heat transport also were assessed. A principle finding was that cyclic heat loads tended to aid startup of the loop heat pipes at the low power inputs.

In this paper we present a hybrid Sliding Mode Control (SMC) and Constant Amplitude Feedback Control (CAFC) architecture for slewing flexible structures. The SMC controller is used to servo the flexible structure and the CAFC is used to suppress residual vibrations of the flexible structure. A single-axis encoder/DC motor/harmonic drive assembly is used for large angle slewing. A graphite/epoxy composite structure with embedded strain sensors/actuators is used for active vibration suppression. The results of this study include the analytical dynamic and control system development with experimental verification. The hybrid control algorithm uses the output sensor data from the encoder and strain sensor along with filters to derive velocity information to compute the control effort for the motor and strain actuators. Near-minimum time maneuvers based on an equivalent rigid structure are used to slew the flexible active structure. The tip mass was varied to evaluate control system robustness. Experimental slewing studies were performed to compare the benefits of using active rather than passive structures. For the active case the experimental results showed a reduction in residual vibration and settling time.

In this paper, we summarize the research being carried out by the author and his group of graduate students on new control and observer design methods specifically tailored towards to control of smart and flexible structure applications. The new approaches are tailored to this specific application in the sense that they take into account the various types of available measurements and provide controllers and observers in terms of the various combinations of measurement variables. This approach of tailoring the controller and observer structure around the available measurement models has significant advantages over traditional approaches. Specifically these approaches introduce two new frameworks for controller and observer design, namely “Reciprocal State Space Framework” and “Matrix Second Order Framework”. Currently State Space Framework is a popular framework for control systems design in the sense that designers always transform the dynamic model of their application into state space form and then carry out the standard design techniques of this framework. However, we argue that there are many applications in which the dynamics, measurement models and performance constraints are such that it is unnecessary and cumbersome to force this transformation and instead the design can be carried out more easily in these newly introduced frameworks that are compatible with the plant dynamics, measurement models and performance constraints. Using these frameworks, simple control and observer designs are carried out with application to the problem of control of smart, flexible structures.

The accuracy of stretch formed sheet metal parts in the aerospace industry depends heavily on tool designer and machine operator expertise. In addition, fixed-configuration facilities and rigid tools result in long lead times, high manufacturing costs and insufficient product quality. In response to this, a methodology combining an adaptive closed-loop shape control algorithm and a reconfigurable forming tool has been developed and implemented for stretch forming. This method is based on empirical estimation of the process characteristics using a “Deformation Transfer Function” and is capable of rapidly generating a tool shape that produces the desired final part shape, even when significant process uncertainties exist. The same control system is also used to compensate for shape distortions caused by the downstream manufacturing processes of chemical milling and trimming. Experiments involving various compound curvature skins are presented that validate the effectiveness of this methodology in reducing the shape errors below the forming system error threshold in just one or two closed-loop forming trials regardless of skin shape.

The looper system, located midway between adjacent stands, is a mechanical levitation device commonly applied on modern hot strip mills. In order to fully understand factors affecting the looper’s behavior and predict its motion during rolling processes, a comprehensively mathematical simulation model of the looper control system for a hot strip tandem mill is developed. The simulator includes three system modules: the looper regulator module, the looper dynamics module, and the mass flow module. Simulation results demonstrate good compatibility with position measurement of real rolling processes. Looper responses under circumstances of different original strip lengths and a varying roller speed are specifically examined. Based on this looper simulator, valuable information can be easily obtained for possible improvements on rolling processes in the future.

This work aims at improving the measurement accuracy of arc length using the spectrum of arc light at a particular wavelength during gas tungsten arc welding with argon shield. To this end, effects of welding parameters on spectral distributions were studied. To verify the effects of base metal and arc length, the arc column was also sampled horizontally as layers for spectral analysis. Results showed that spectral lines of argon atoms are determined by arc length and independent of welding parameters other than the arc length. Based on these findings, a compact arc light sensor has been designed to measure the arc length with an adequate accuracy. A closed-loop arc length control system has been developed with the proposed sensor.

A unique system is introduced for monitoring the viability of conventional position switch interlocks on movable barrier guards. At closely spaced intervals a test device motor causes the guard to open slightly to check if the interlock will change state in an attempt to interrupt control power to the protected machine. The machine will not in fact shut down during the testing phase because the interlock system will be isolated from the machine operating motor contactors during the test duration. Testing of a guard is annunciated just before and during the automatic testing phase. Any failures of an interlock or its testing system will also be annunciated until their causes are corrected. When a latching system is used in conjunction with the guard, a faulty interlock will cause the guard to be locked closed until it is convenient to effect a repair. Production is not interrupted because the interlock remains bypassed during this period. If an interlock is temporarily circumvented or if the interlock guard is sabotaged, the test device motor will not cause a change in state of the interlock during the test phase; again, system malfunction will be annunciated and/or the guard will remain latched and/or the machine will be shut down. The proposed process provides the following advantages: • A quantum leap in system reliability and safety is obtained using even the most ordinary “limit switch” type interlocks. • The tamper-resistance of all known interlocks will be greatly enhanced; all popular forms of temporary interlock circumvention are eliminated. • Long term sabotage which involves simultaneous guard removal and interlock bypassing is addressed for the first time; to defeat the proposed process involves the most outrageous restructuring of the control system. • The system will safely endure adhesive environments that may cause interlock activators to stick in their “fail to danger” states. • The monitored machine is almost never shut down nor production interrupted during or after the testing phase even if mischief is uncovered. • The testing of large numbers of interlocks will not expose the machines’ electric power motors to the frequent restart demand associated with some conventional manual testing procedures. • Without compromising safety, maintenance of even multiple interlock failures may be deferred until their repair is convenient. • The assurance of working interlocks may enable OSHA to expand applications where maintenance is performed under “control stop” as opposed to the more elaborate “lock out.” • The design of the new test and monitoring system may be executed and implemented using straightforward hardware and software that are firmly entrenched technology.