Henry R. Jex

Measurement of Driver-Vehicle Multiloop Response Properties with a Single Disturbance Input

Multiloop response properties of controllers are, in general, very difficult to obtain because an independent forcing function is needed for each describing function to be measured, and interpolation procedures may be required to obtain intermediate describing functions at common frequencies. Even then, a certain amount of untangling is required before the final results are obtained. When the loops that are closed and the nature of the describing function forms adopted in each loop are known or hypothesized, matters can be made much simpler. Then, the quantitative values of the individual describing functions can be readily identified using appropriate closed-loop describing function measures and decomposition procedures. Two examples are provided for the measurement of driver-vehicle multiloop response properties using a single disturbance input. The validity of the procedure is based on current multiloop operator adjustment rules and is made plausible by comparison with experimental data.

Defining and Measuring Perceptual-Motor Workload in Manual Control Tasks

“Everyone knows” when he is subjected to a high mental workload or stress in a complex manual control task such as driving a car, steering a ship, or flying an airplane. Nevertheless, there is, as yet, no accepted definition of operator workload. This is mainly due to the incommensurate dimensions of various loading tasks and the lack of any comprehensive theory or validated models.

DRIVING SIMULATION-REQUIREMENTS, MECHANIZATION AND APPLICATION

This paper discusses recent developments and application of driving simulators. Modern simulator requirements are reviewed from the point of view of both driver characteristics (vision, audition, proprioception, vestibular motion seeensation) and task demands (e.g., steering and speed control, risk perception, decision making, general workload level). A variety of simulator applications are summarized, including comparison with subsequent field tests. These applications include studies involving drunk driving and risk taking, reduced visibility and delineation, and signing. Possible future simulator developments and application are also discussed.

DEVELOPMENT AND CALIBRATION OF AN AERODYNAMIC DISTURBANCE TEST FACILITY

This paper summarizes the development of an aerodynamic disturbance test facility that can be used to study the influence of crosswind disturbances on road vehicle handling. A literature review of ambient wind measurements and wind tunnel tests was combined with preliminary analysis to develop the requirements for a low cost, transportable, easy to operate, rugged, test facility.

Control Characteristics of Alcohol-Impaired Operators

Although the operation of vehicles like airplanes, cars, and bicycles involves a complex array of perceptual, decision and control activities, most accident statistics clearly show that intoxicated operators are a dominant cause of accidents, and not the difficulty of the task itself. This paper summarizes some recent research on the nature of the impairment of operator control under blood alcohol concentrations (BAC) up to above 0.16 percent. Alcohol toxicity is shown to be quite specific with respect to visual-motor functions involved in control of a vehicle, and experiments with a generalized workload task and special driving simulator show how these are reflected in terms of changes in operator control parameters such as response latency, gains, stability margins, and coherency.

LTASIM: A Desktop Nonlinear Airship Simulation

A comprehensive nonlinear 6 DOF simulation model for modern airships has been developed (LTASIM) using the TUTSIM-6 program for use on PCcompatible computers. The model includes nonlinear aerodynamics and buoyancy effects, various thrustor types, important virtual mass and inertia tensor terms, non-constant wind effects, stability and trim controlsystem, airship-to-target relative motions. On a fast 25MHz 486-type computer, it will run much faster than the simulated time interval, thereby offering possibilities for use in real-time simulation devices.

DRIVER DECISION-MAKING RESEARCH IN A LABORATORY SIMULATION

This paper reviews a simulation approach to the study of driver decision-making behavior. Tasks were set up with rewards and penalties applied as performance incentives. Simulation tasks were selected that would fit into a decision context and could be efficiently implemented. Familiar driving situations were chosen to give an interesting, motivating driving scenario and cover a range of decision-making situations. The tasks were programmed to occur randomly and repeatedly in a simulated continuous drive. Expected Value Theory was used both to set up task conditions and as a means for data analysis and interpretation. Both performance and decisionmaking behavior were measured. Decision-making behavior was consistent with an Expected Value decision-making model interpretation.