U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-95-153
Date: November 1996
Development of Human Factors Guidelines for Advanced Traveler Information Systems and Commercial Vehicle Operations: Literature Review
THE IMPACT OF EMERGING TECHNOLOGIES ON ATIS/CVO SYSTEMS AND HUMAN FACTORS DESIGN CONSTRAINTS
The purpose of this section is to provide an overview of applicable emerging technologies for ATIS/CVO applications. Much of the technology of today and tomorrow is being focused on providing drivers with information on road conditions, navigation, warning systems, vehicle controls, and personal communications. In order for information to get to the driver, it must be sent to the automobile, interpreted, and displayed. This section covers current, near–term, and future forms of communication devices, displays, and navigation technology that have the potential to impact the human factors design of ITS systems.
As more motorists take to the road and driving tasks become automated, drivers need a way to communicate with a system to optimize travel time and safety. Citizens band and amateur radio are outdated due to the need for faster and more reliable devices that can also transmit and receive data. Today's mediums of data and voice communications include infrared, frequency modulation (FM) sideband, mobile–satellite services, cellular, radio frequency (RF) data networks, inductive–loop systems, and Shared–Trunked Radio Systems (STRS) (Weld, 1989; Kirson, 1991).
Infrared systems use roadside beacons to transmit and receive information to and from equipped automobiles. They provide an excellent rate of data transfer and have a low cost. However, they must be in proximity to the car and environmental conditions can disrupt the signal.
Infrared beacons could be used to support AVI systems with either one– or two–way communications. Beacons can also be used as navigational aids. The beacons can update an automobile's position on the map data base as the car passes by. Information about upcoming intersections can also be provided.
FM sideband technology takes advantage of sideband radio and TV frequencies and broadcast information. This format is inexpensive and requires no additions to the automobile. The United States has used a highway advisory system on the amplitude modulation (AM) dial since the 1970's. Several European companies have developed more complex systems that broadcast a code at the start of the message so that only cars affected by the information will receive it. Other advances allow drivers to listen to non–critical information at their leisure and have critical information mute their radios or tape players (Davies, Hill, and Klein, 1989). These more advanced systems require a device to decode and present the information. Possible display formats include in–dash information displays and speech synthesis. Usability of this system is limited since it provides only one–way communication and some areas are not suitable for receiving FM transmissions.
One possible short–term use of sideband technology would be to provide up–to–date traffic information to all drivers in a local area. For example, units could be sold in varying degrees of complexity. The low–end model would intercept all information and display it on a small monitor. More expensive models could use coded signals to present information that is only relevant to the driver in the area. These systems would be ideal for the traveler who does not need route guidance or trip planning, but needs to know current traffic and road conditions.
Mobile satellite services are advantageous because they transmit and receive information directly to and from an automobile regardless of geographic location. Relative costs for satellite systems are low and transmission speeds are high (about 2400 bits/s). A disadvantage of satellites is that they require several cities to use the same channel, which limits the total usage.
The next major milestone in communications will be the capability to transmit and receive voice or data from land, air, or sea from hand–held terminals. The iridium system is composed of 77 low–earth–orbit satellites. A major advantage of this system is the ability to switch between the satellites as the user moves across the country with no loss of data between links. The satellites will be able to simultaneously cover all populated areas of the earth. Motorola plans on having all 77 satellites in orbit by 1997, with initial launchings in 1995. Their primary use will be to back up terrestrial stations of cellular communications. Because the area of the earth that each satellite will cover at any one time is large, each satellite would not be able to handle as many calls at once as a ground station. For IRANS, the satellites will be able to track a user with much greater accuracy than the current Global Position System (GPS). A possible shortcoming of the system could be a lack of cooperation between cellular companies. Since it would be possible for anyone to "dial direct" to a satellite, local dispatches would have to regulate calls and provide billing.
Cellular technologies use land–based centers, each with several cells capable of transmitting information. A mobile unit uses the strongest cell to communicate and can be handed off to another cell when a stronger one comes into range. Newer digital cellular links are being formed that will improve the reliability and transfer rate of information. Approximately 2 percent of Americans use cellular technology and cells are already becoming overloaded. This problem, plus the fact that many areas are not cellular equipped, suggests that cellular technology as it exists today will not be beneficial to large–scale ATIS uses. In order for cellular communications to be a useful source for communications in the future, cells will have to accept more users at one time and will need a greater range.
Radio Frequency (RF)
RF data networks may prove to be an expandable resource for ITS applications. The system operates in a manner similar to the cellular communication links, but does so at a much lower cost. This cost is lower on both the user's and the transmitter construction ends. Each new location of coverage requires its own antenna. RF technology has proven to be successful. An application of radio transmission, called Packet Access Radio (PAR), uses short spikes of data sent through either normal RF nodes or bounced off satellites or meteor scatter (Williams, 1989). Williams claims that meteor bounce could be an excellent low–cost data communication medium. The biggest drawback of PAR is its transmission speed; some transfers can take several minutes. For non–time critical information, meteor bounce should be investigated further for ITS communications.
Inductive–loop systems are mounted under the roadway surface and are mainly used to track and detect vehicles. An alternate use would be to allow communications between the loop and the automobile. The major drawbacks of this system include low data rate and a range limited to the length of the loop. Installation costs are also high, since each loop must be buried under the road surface.
ATIS/CVO uses of this technology include AVI and areas that constantly change status. For example, commercial drivers entering an area may need to follow a specific path to avoid dangerous areas. Loop systems could be used to guide the vehicles in the right direction and inform a control center when a vehicle has entered a dangerous area.
Shared–Trunked Radio Systems
Shared–Trunked Radio Systems (STRS) operate in the same way as cellular radios, but use a 300–MHz band. Mobile units either lock onto a control channel or scan channels available for transmission. The major difference between cellular and STRS is that cellular units need enough dedicated cells to cover all users, where STRS users share cells that are not being used. STRS only covers about 30 percent of the United States, and adding more transmission units would be expensive.
Regulations determining ownership and responsibility of areas and bands of communication are important communication concerns. In order for any of the above systems to work, an agreement must be reached between existing companies and the Government to provide data communications standards (Chadwick and Patel, 1992).
In–vehicle data transfer
The current and traditional method of getting information from one part of the automobile to another is wrapped metal wire. At least two new methods have been proposed: plastic optical fiber and Spread Spectrum Transmission (SST).
Optical fibers have proven to be very effective in sending data at high speeds. The typical fiber used today is glass, which is more expensive and more fragile than plastic. Plastic fibers, when stretched over several miles, do not have the same performance as glass fibers. However, for the relatively small confines of an automobile, plastic fibers offer excellent data transfer at a low cost and weight. Another advantage of plastic optical fibers is that when encased in a molded shell, they can withstand the rigorous environment (including shock and temperature) of the automobile in transit. Advances in connections have also made plastic fiber easier to work with and install (Ueda and Yamaguchi, 1990). Plastic optical fibers can also be used in a passive network system that allows shared system information within the car to be received by all systems at the same time with only one transmission (DiLiello, Miller, and Steele, 1990).
SST is a wireless system used to connect local area networks (i.e., several computers sharing the same hardware). SST has been used by the military for several years, but its specifics were not known until recently. The greatest advantage of SST is its ultra–high security and accuracy in data transfer. This is accomplished by diffusing a radio signal over a large band and then decoding the signal at the other end. Any attempts to intercept the signal with normal radio interception will only get part of the spread signal. Uses of SST technology are widespread, from inter–automobile communication to networks of ATIS support stations. On–board computer systems could be wireless. The system is already being used for sending real–time data from test vehicles to portable computers on site. As of 1989, only one company had received Federal Communications Commission (FCC) authorization to use SST without a radio license (Zenko, 1989).
In–vehicle display requirements
As automotive–compatible technologies increase, the need to present information quickly and saliently has become more important. According to Akiba, Davis, Kato, Tatiyoshi, Torikai, and Tsunesumi (1991), the automotive requirements for displays are:
The major categories of current displays are Vacuum Fluorescent Displays (VFD's), CRT's, LCD's. Those for near–term and future displays are HUD's, and Helmet/Head Mounted Displays (HMD's) respectively.
Vacuum Fluorescent Displays (VFD's)
VFD's are currently the most common type of displays found in automobiles. They are used in clocks, digital speedometers, message centers, audio systems, and temperature control systems.
VFD's are produced by exciting a phosphor–coated anode. The current colors available are blue, green, yellow–green, green–yellow, yellow–orange, orange, and red–yellow. With filters, the color combinations increase. VFD's provide high luminance at low voltage cost, are highly readable, and have a life span of more than 10,000 h. Recent advances in VFD technology have increased display area size, created a greater range of colors, and reduced the voltage use to half–duty cycle. Future advances will provide full–color displays larger than the current 127 by 178 mm (5 by 7 in), use graphics, increase luminance for HUD technology, and will run at a one–third duty cycle. See Akiba et al. (1991) and Iwasa (1991) for more detailed information on VFD's.
Cathode Ray Tube displays (CRT's)
CRT's are now being used in some automotive applications. CRT's are more commonly seen in computer displays and television screens. They can present full colors; have high resolution; and because of their maturity, are currently the least expensive of the major display technologies. The major disadvantages of CRT's for automotive use are their large size, weight, and power consumption. Another disadvantage is that as the screen size gets bigger, the image gets dimmer, a critical factor in the glare–ridden vehicle environment.
Liquid Crystal Displays (LCD's)
LCD's are currently receiving the most display research attention. According to Nordwall (1989), "LCD's generally offer savings of about 60 percent in volume, 70 percent in weight, and 80 percent in power compared with cathode ray tubes." LCD's can also display color using built–in filters. However, LCD's are non–emissive, have a narrow viewing angle, and some types have difficulty operating under high and low temperatures (Erskine, 1988).
There are several different types of LCD's available. The current type used in automobiles today is the Twisted Nematic (TN) configuration. In its simplest form, an LCD works by applying voltage to a "sandwich" cell consisting of a polarizer, a glass substrate, a transparent conductor, and an alignment layer on either side of the twisted nematic liquid crystals. When voltage is applied to the cell, the conductors cause the alignment layers to "untwist" the liquid crystals. When this happens, the polarizers line up and let light through. An obvious disadvantage of LCD's is brightness. With this basic TN configuration, the more cells (pixels) you have, the more voltage it will take to drive them. If too much voltage is applied to a row or column, other unwanted pixels may open (called "bleeding"). A second disadvantage is that it takes a set amount of time for the voltage to "spread around" the network of cells. One solution to these problems is the use of "switches" on each pixel that can either allow the appropriate energy through the cell or store the charge until the cell is needed. These switches are most commonly Thin Film Transistors (TFT) or Thin Film Diodes (TFD). This type of display is called an active matrix LCD. A good general description of LCD's and active matrix LCD's is presented in Firester (1988). A more recent type of LCD, called a Double–Layered Super–Twisted Nematic Liquid Crystal Display (D–STN LCD), has also been developed. Its major benefits are a wider range of temperature operation, lower voltage usage, and increased contrast ratio (Matsumoto, Nakagawa, and Muraji, 1991). D–STN LCD's can also be manufactured at a lower cost than other active matrix displays (Itoh, 1991).
Although the technology exists to create large flat panel displays, they are expensive. Close to half the cost is in the fabrication process. Much of the current research today is focused on creating cheaper and larger panels on which to build displays (see study by Takeda, Ezawa, Kuromaru, Kawade, Takagi, and Suzuki (1989), for an example). A large flat–panel LCD would allow one large panel to replace the several smaller gauges and dials in most current automobiles. These displays could be made the same size for each manufacturer's models and could be programmed to have different appearances. The displays could also include the displays users want to see.
Since LCD's are non–emissive, they need an external source to light them. The two most common types of back–lighting lamps are the Cold–cathode Fluorescent Lamp (CFL) and the Hot–cathode Fluorescent Lamp (HFL). The HFL provides a higher luminance than the CFL, but has a higher operating temperature and shorter life (i.e., approximately 3,000 h). A newly developed lamp is the Warm–cathode Fluorescent Lamp (WFL). The WFL offers twice the intensity of the CFL and 10,000 h of operation equal to that of the CFL. The WFL also operates at a lower voltage level and can be constructed with a thin film heater to help the liquid crystals operate better at temperatures below 20° C (68° F) (Itoh, 1991).
Head–Up Displays (HUD's)
The HUD can use one of several projection sources to project an image (e.g., mi/h, warning indicators, etc.) onto the windshield. This information appears to be floating in space in front of the vehicle. The HUD allows drivers to keep their eyes forward, without glancing to the dashboard. HUD's have been successfully used in aircraft by giving the pilot a "window" to fly through. An automotive HUD is different from an avionic HUD in that the scenery behind the display is more complex for the driver than for the pilot. A second difference is that automotive HUD's are displayed not at optical infinity, as in an aircraft, but at a closer distance, somewhere between 1.8 and 7.3 m (6 and 24 ft) (Stokes, Wickens, and Kyte, 1991).
HUD's are commonly produced by either reflecting an LCD off of the windshield by means of a half–mirror or direct reflection, or by using a light source to illuminate a holographic element on the windshield. Proponents of both systems claim success with each of these systems, and much research is currently being conducted to create the best optical picture at the lowest cost. For detailed information, see Patterson (1988), and Wood (1988). A high–luminance VFD has also been proposed as an alternate HUD projector because it needs no back–lighting and is resistant to shock and temperature conditions.
Helmet/head–Mounted Displays (HMD's)
HMD's are currently used in military aircraft. The current technology places a HUD on a monocular "tube" that is clipped onto a helmet and placed directly in front of one of the subject's eyes. The advantage of having a HUD on your head is that it moves with you as you pivot your head. The AH–64 Apache helicopter uses a monocular HMD for night flying and targeting. In the near future, HMD's will cover both eyes.
A second type of HMD uses a small CRT display mounted on a helmet that fills the field of view of the pilot. The image can be either computer–generated or manipulated real–time video footage of the outside world. For example, a helicopter pilot could land a craft in rough terrain by "looking" via the floor through a remote camera.
It is probably safe to assume that in the near future, the typical driver will not want to wear either system of HMD in normal driving situations. However, the CVO driver who needs to travel in low–visibility conditions could use a monocular HMD with a night–vision device to navigate to a delivery location. Law enforcement officers could also benefit from night–vision HMD's. Navigation and warning controls could also be put on the display. It is feasible that monocular displays could be put on a hat or some type of glasses so they would be easy to put on and take off.
According to French (1988), there are three basic types of navigation systems. Autonomous navigation systems are capable of operating without the need for external sources such as satellites or road beacons. Most autonomous systems use dead reckoning to track the distance and angles traveled. Dead reckoning uses wheel rotations and directions turned to estimate the current position of the car. The Japanese Multi–Advanced Vehicle (AV) system uses a new optical fiber gyroscope to more accurately sense vehicle turns (Oshishi and Suzuki, 1992; Harrell, 1991). Dead–reckoning systems have historically been relatively reliable, but periodically get out of calibration. The result is that the driver is given incorrect information about location and/or route status. Therefore, provisions for out–of–position information and simple location adjustment are necessary. These features use route–map displays.
Radio navigation systems use satellites to keep track of an automobile's position. Each vehicle to be tracked has a unique code that can be "seen" by the satellite and reported to a base station. The most commonly used satellite is the GPS, but there are many companies around the world competing for the market. Most systems that use radio navigation also use dead reckoning to account for areas where signals may be blocked. Currently, the GPS system of satellites is incomplete over the United States. In addition, satellite signals are often blocked by tall structures in major cities. Therefore, it appears likely that many ITS systems will always require backup navigation systems if GPS is used.
Proximity beacon systems use short–range transmitters to send signals to passing cars with receivers. These beacons are also usually combined with dead reckoning and satellite signals. The beacons have a correction factor that updates position from the other methods. A drawback to beacon technology is its cost of construction and maintenance.
The most common method of presenting information to drivers with navigation systems is through video monitors that display a map of the area and the automobile's current position. Some systems also use voice synthesis to convey messages to the driver. Much research needs to be conducted to determine which properties of voice are most salient in the driving situation. The Back–Seat Driver, a Massachusetts Institute of Technology (MIT)–based navigation system, uses only speech to guide drivers (Davis and Schmandt, 1989). Davis also provides a good explanation of some potential problems with speech–based directions.
Synthesized speech is not as intelligible as digitized speech (Marics and Williges, 1989). Intelligibility is particularly important in vehicle environments due to the high noise levels often present. It may be some time before the intelligibility of low–cost synthesized speech systems required for in–vehicle use are of any quality. Unfortunately, for many ITS applications, synthesized speech is desirable because large data bases are sometimes required for communication of certain information (e.g., next street name in a large city) (Dingus and Hulse, 1993).
Motorola has developed a voice recognition system that will be available in automobiles. The system is user–dependent (i.e., a user must "train" the system prior to use) and a new driver cannot effectively use the voice commands. The reliability of this system in the automotive environment is unknown; however, it is apparent that advances in speech recognition technology could make this control technique viable for ITS applications.
The information presented in this section is the result of a broad literature review. Because technology advances at such a rapid pace, this report is by no means comprehensive. The technologies reviewed are the next advances that will probably occur. Specifically, LCD technology, digital cellular voice and data communications, and speech synthesis and recognition may impact ATIS/CVO system design in the near to middle term.