Search Results

The concept of a Variable Dynamic Testbed Vehicle (VDTV) has been proposed as a tool to evaluate collision avoidance systems and to perform driving-related human factors research, among others. The goal of this study is to analytically investigate to what extent a VDTV with four-wheel-steering can emulate the lateral dynamics of a broad range of vehicle models. Using a particular mid-sized vehicle as a baseline, our study indicated that this mid-sized vehicle modelled with a closed-loop four-wheel-steering system can be controlled to emulate the lateral response characteristics of a range of vehicles, from “small” to “full-size,” reasonably well over a speed range of interest. A novel steering control configuration that has the potential to improve further the “degree of emulation” has also been proposed.

A variable stability test bed is under development for the National Highway Traffic Safety Administration (NHTSA). The Variable Dynamic Testbed Vehicle (VDTV) is being designed for research and testing of advanced collision warning and avoidance technologies being developed by industry and most likely being made available to consumers in the near future. The VDTV will also be used by NHTSA in support of the Automated Highway System (AHS) Program and possibly by the AHS program directly. The VDTV will have advanced dynamic subsystems that can be varied by on on-board programmable computer Suspension, steering, throttle, and braking will thus be controlled through selected algorithms that may be changed to provide a reasonably broad range of vehicle dynamic characteristics. The vehicle is inherently a drive-by-wire system, is instrumented for both vehicle and human factor measurements, and is therefore ideally suited to many Intelligent Transportation Systems (ITS) applications.

The Variable Dynamic Testbed Vehicle (VDTV) is presently being developed by the National Highway Traffic Safety Administration (NHTSA). It is being designed to have a “steer-by-wire” front steering system and an independent rear steering system. These steering systems enable the VDTV to emulate the directional control characteristics of a broad range of passenger vehicles. In this study, a “model-following” control method is used to modify both the steady-state and transient lateral response characteristics of a small-size VDTV to match those of compact-size and mid-size vehicles. For two classes of steering inputs considered in this study (“pseudo-step” and “sinusoidal”), the model-following control design method allowed the VDTV to accurately and robustly track the lateral responses of the target vehicles over a range of forward speed.

The Variable Dynamic Testbed Vehicle (VDTV) concept has been proposed as a tool to evaluate collision avoidance systems and to perform driving-related human factors research. The goal of this study is to analytically investigate to what extent a VDTV with adjustable front and rear anti-roll bar stiffnesses, programmable damping rates, and four-wheel-steering can emulate the lateral dynamics of a broad range of passenger vehicles. Using a selected compact-sized automobile as a baseline, our study indicated this baseline vehicle can be controlled to emulate the lateral response characteristics (including the vehicle's understeer coefficient and the 90% lateral acceleration rise time in a J-turn maneuver) of a fleet of production vehicles, from low to high lateral acceleration conditions.

This study addresses the performance of four-wheel-steering vehicles in high-speed lane change maneuvers. We first compare the recorded steering command of an experienced driver in executing a lane change maneuver with that determined via solving a suitably formulated optimization problem, and found them to be qualitatively comparable. This finding allows us to analytically compare the optimal lane change performance achievable with both two and four wheel steering vehicles. For a representative high-speed lane change maneuver, our study revealed that, in the hands of an experienced driver, the performance benefit achievable with four-wheel-steering vehicles is not significant. This conclusion is in agreement with road-test results obtained with two production four-wheel-steering vehicles. The applicability of the proposed “optimal control” approach in evaluating the performance of other driver-vehicle maneuvers is an interesting topic for future study.

A “lane change crash” is defined as a family of collisions that occurred when a driver attempts to change lane and strikes or is struck by a vehicle in the adjacent lane. One type of maneuvers that is commonly used to avert a lane change crash involved aborting the intended lane change, and returning the vehicle to the original lane of the subject vehicle. This study addresses the performance of driver-vehicle systems in aborted lane change maneuvers. We first compared the recorded steering command of an experienced driver in executing a lane change maneuver with that determined via solving a suitably formulated optimization problem, and found them to be qualitatively comparable. This finding allows us to analytically assess whether an experienced driver can successfully avert a lane change crash if he responsed to the warning from a collision detection system Te seconds after the initiation of the lane change maneuver.