Evaluation of Camera-Based Systems to Reduce Transit Bus Side Collision
Many transit bus crashes in which the bus is at fault are sideswipes caused by the drivers' inability to see objects or other vehicles in the blind zones of the mirrors on the bus. This research compared relatively new camera-based systems to sensor-based and mirror-based systems for transit buses for their potential for side collision avoidance and their reliability in rain and at night. The camera-based system for transit buses under evaluation was a system that incorporates video cameras installed on the outside walls of the bus, aimed at the left and right rear sides of the bus, and two monitors connected inside the bus to provide the driver with an image.
The volumes of blind zones for mirror- and camera-based systems were measured for three types of buses: the Ultra-Low Floor design commonly used by transit agencies, the high floor design used for school buses, and the cutaway design often used for shuttle or paratransit services. To perform the measurements, a bus driver sat in the driver's seat with the mirrors and seat adjusted according to his height while communicating via two-way radio with a team member who was walking outside the bus and checking if various locations were in the driver's field of vision.
During the controlled driving test, 28 drivers using the ultra-low floor and cutaway buses were given time to practice using the camera system while on a course delineated in orange cones and containing left turns, right turns, and a lane change. The drivers' distance perception with the mirrors and cameras was tested in both static and dynamic conditions. Static conditions involved estimating the coordinates of a stationary person standing behind and to the left or right of the bus. Dynamic conditions testing required the bus driver to indicate the last safe moment to make a lane change while the bus was cruising straight at 15 mph with another vehicle approaching the bus from behind in a parallel lane.
Drivers were given surveys before and after the driving tests. Also, feedback was obtained from the drivers about possible setup locations, monitor sizes, and orientation, as well as enhancements to make the sideview video system more attractive for use.
- The camera-based system with a regular angle lens reduced 43 percent of blind zones compared to a common combined flat and convex mirror system, and wide-angle camera systems were able to entirely eliminate blind zones. Most drivers adapted to the cameras quickly. In the static and dynamic distance perception tests, drivers were able to make similar distance estimates using the camera- and conventional mirror-based systems. The authors further found that the cameras perform better in rainy conditions and, because of infrared sensors, better at night than mirrors; however, in certain lighting conditions, the cameras experienced certain limitations.
- Eighty-two (82) percent of the bus drivers agreed that "the sideview video system can improve bus safety." While only 46 percent of the drivers would "agree to replace the mirrors with the sideview video system," 68 percent of drivers agree that they would "like to have the sideview video system in the bus [they] drive every day." The authors emphasize the camera system used for the research was an after market product and that deployed systems could still be greatly enhanced.