Currently, there is a performance issue regarding vehicle control at higher speeds for some indirect-vision, by-wire military vehicles; that is, those vehicles in which mechanical links between the driver and control devices are replaced by electronic or computerized signals. Work has been performed to assess the current state of knowledge regarding the shaping function. The overall goal was to identify design parameters critical to improving the current by-wire implementation for military tactical vehicles, and to ultimately optimize system (i.e., human-vehicle) performance for the execution of secure mobile operations.
Future military vehicles will be drive-by-wire (DBW) vehicles, which means that mechanical elements of the control system are replaced by operator-controlled input devices coupled with remote actuators via a central electronic control system. A DBW vehicle can be composed of several “x-by-wire” subsystems. Examples include steer-by-wire, brake-by-wire, and throttle-by-wire, each referring to the individual vehicle control subsystems for which direct mechanical linkages (such as the hydraulic brake line) have been replaced by electrical signals between the input device (e.g., brake pedal) and the actuators of the system (e.g., calipers). This work focuses on steer-by- wire subsystems.
Because the actuators in x-by-wire systems are controlled by electrical signals from a computer rather than a direct mechanical link to the human-machine interface (HMI), input from the driver can be supplemented with or modified by intelligent automation. A good example of such automation in civilian vehicles is adaptive cruise control (ACC). An ACC system is used in a manner similar to standard cruise control, except that after the driver sets the desired speed, an adaptive controller regulates spatial separation from other vehicles on the road as well as attempts to maintain the speed indicated by the driver.
For the purposes of this research, a shaping function is defined as a mathematical description of the scaling between the input and output of a given system; that is, how the input is “shaped” into output. This is similar to the concept of a transfer function, except that the transformation is not explicitly occurring in the frequency domain. For steer-by-wire systems, the shaping function maps the angular displacements of the HMI control input device (joystick, yoke, or steering wheel) to the system response in terms of the vehicle steering angle.
The examination of shaping functions is important for several reasons. First, the use of HMI devices with limited “throw,” or total angular range of displacement, will pose challenges to the operator. A specific consequence of limited throw is the magnification that the operator will perceive relative to his or her expectations, based on experience with standard vehicles with steering wheels. In other words, because of the smaller permissible angular range of alternate HMI, there will be a lower input position-to-wheel angle ratio (also known as the steering ratio), and therefore, smaller hand/arm motions will produce larger vehicle responses.
A second issue involves the need for differential steering sensitivity across various driving tasks. Consider, for example, the large angular range of steering motion required for parallel parking as compared with the relatively small motions needed for lane maintenance while one is driving on a highway. Now envision the variety of steering tasks that may be encountered on the battlefield or during off-road missions in military vehicles. Such task factors are further compounded by the third issue: vehicle speed. In particular, as the vehicle moves faster, the magnitude of lateral accelerations during steering increases, elevating the risk of oversteering, inducing excessive roll or loss of control leading to collision, spinout, or rollover.
An advantage to the electronic control used in steer-by-wire is that the steering ratio does not have to be fixed, but rather, it can be modulated adaptively across platforms, tasks, and speeds to help account for these performance and safety issues. The primary approach to overcoming the problems mentioned has been to implement variable gear ratio (VGR) steering systems. The gear ratio in a steering system provides a description of how much angular displacement of the steering wheel is required to produce a particular angular displacement of the vehicle heading via turning of the wheels. For instance, if one complete revolution (360 degrees) of the steering wheel results in the wheels of the vehicle turning 20 degrees, then the steering ratio is 360/20, or 18:1. A higher ratio means that one has to give greater input (thus expending more energy) to get the wheels to turn a given distance.
The use of a shaping function can eliminate the need for or facilitate the function of an on-line, dynamic control system that monitors vehicle motion and changes the steering ratio from one speed condition to another. That is, if one considers that high-speed driving is most commonly associated with a very small range of steering input and lower speed maneuvering (such as parking) uses much more of the dynamic range of the steering wheel, then a nonlinear shaping function that applies a high ratio for small steering input and a lower, more direct ratio for large input will effectively match the steering ratio to vehicle speed by virtue of the task that the driver is performing. At the same time, one can envision dynamic selection of shaping functions that are optimized for particular vehicle operation scenarios in order to smoothly change steering characteristics in an on-line fashion when a different vehicle response is required.
When one is considering performance difficulties in HMI, the interface should be one of the primary system elements assessed as a potential source of error. The functional capacity of the driver-vehicle system will be defined in large part by the interaction between the capabilities of the operator and the operational characteristics of the HMI. By-wire systems provide a myriad of possibilities in terms of the HMI devices that can be used for vehicle control. This flexibility, combined with a general lack of complete understanding of human factors issues associated with these interfaces, creates a level of ambiguity with respect to the establishment of design standards regarding ideal operational characteristics for steer-by-wire vehicles. Careful design decisions are needed since driver performance remains compromised at higher vehicle speeds and when nonstandard HMI devices, such as joysticks, are implemented.
The results of the experimental studies that have been completed thus far are providing important information specific enough to facilitate the establishment of design guidelines for intelligent military vehicles.
This work was done by Jason S. Metcalfe of DCS Corporation; and Susan G. Hill and Kaleb McDowell of the Army Research Laboratory. ARL-0053
This Brief includes a Technical Support Package (TSP).
Using a Steering Shaping Function to Improve Human Performance in By-Wire Vehicles
(reference ARL-0053) is currently available for download from the TSP library.
Don't have an account? Sign up here.