Smart Buses, Smart Intersection
Shine at Washington IVI Meeting
by Ching-Yao Chan,
Jim Misener, Joanne Lins, California PATH

California PATH staff demonstrated the results of three current research projects at the National Intelligent Vehicle Initiative Meeting held in Washington, DC from June 24-26, 2003. The technically successful demonstrations, held at the Turner-Fairbanks Federal Highway Administration (FHWA) Research Center, in McLean, Virginia, won high praise and contributed significantly to the IVI meeting. They illustrated the professionalism and technical capabilities of PATH teams, and promoted awareness of PATH among peers and transportation organizations around the world.

Intersection Decision Support

One demonstration introduced an intelligent intersection, the product of the Intersection Decision Support (IDS) project sponsored by the FHWA and Caltrans. PATH demonstrated a promising, near-term deployable IDS system that warns drivers when it is unsafe to make a permitted left turn in the face of an oncoming vehicle. Using multiple detection and sensing devices (including Lidar, radar, inductive loop detectors and in-vehicle GPS), the system can identify and track vehicles approaching the intersection in real time. A central processing unit (CPU) fuses the vehicle motion data from the sensors with the signal timing and phasing data sent from the intersection's traffic controller to run a decision making algorithm. When conditions for making a left turn are unsafe, the system triggers a large flashing "No Left Turn" road sign to warn drivers of a hazard. The circle/slash under the "No Left Turn" arrow not only flashes, but grows 50 percent in size and thickness. This apparent "looming" motion affects the fastest and most sensitive pathways in the visual nervous system and makes the sign especially visible. The sign is placed just above eye level at the opposite corner of the intersection. IEEE 802.11a wireless communication is also incorporated in the system, to allow direct communication between the CPU and approaching vehicles, which could provide information directly to in-vehicle devices.

Click Here to see a demo of the Intersection Decision Support Demonstration. (Real Video File)

Frontal Collision Warning System

PATH's second demonstration featured Frontal Collision Warning Systems on a San Mateo Transit bus, a project sponsored by the Federal Transit Administration (FTA). The demo showed how bus drivers can benefit from a warning system that uses radar, lidar, and computers to "watch" the operating environment and a driver-vehicle interface to warn the driver of a possible frontal collision. For several years, PATH has conducted research to understand the causes and consequences of transit frontal collisions. The knowledge gained has led to the development of a prototype collision warning system that provides the driver with an effective warning when the system determines that a potential collision may occur. Based on a well established data fusion model, a preliminary detection algorithm was developed that can track different obstacles within the sensor field of view and can decouple the bus motion from the sensor measurements. A warning algorithm was also developed to incorporate a warning threshold synthesized from the drivers' normal braking behavior. When the system determines that the distance to a vehicle or obstacle in front of the bus is closing too rapidly, it lights up two orange LED lightbars mounted on each side of the windshield. The LED elements are illuminated sequentially, giving the illusion of a moving bar of light. The more imminent the collision, the longer and faster the bar moves. This prototype has been put into regular transit service for field testing, to verify if the performance requirements developed under this project are within a reasonable and reachable range.

Click Here to see a demo of the Frontal Collision Warning System. (Real Video File)

PATH human factors researchers worked closely with SamTrans drivers to understand their needs and expectations and their operational environment, and to define system boundaries. Drivers' inputs and their active involvement in brainstorming different design options greatly contributed to the design of the prototype Driver Vehicle Interface (DVI). Through extensive field tests and close interaction with bus drivers, the system is being improved, and is now being integrated with side collision warning system through a joint effort by a California and Pennsylvania partnership to provide an integrated solution for transit collisions.

Precision Docking

The third demo showcased precision docking, an essential element of Bus Rapid Transit (BRT) operations. The PATH system involves a variety of BRT technologies implemented on a 40-ft compressed natural gas (CNG) bus, made possible by a FTA lane-assist project and a Caltrans BRT development project.

In the precision docking demonstration, the bus steered itself along a demonstration course representing the path a bus would take to approach a curbside bus stop. The bus was completely computer-controlled, automatically following a series of magnets that defined the course, and using a smooth speed profile for its acceleration and braking.

Click Here to see a demo of the Precision Docking Demonstration. (Real Video File)

A 90-meter-long path of magnetic markers was placed on the roadway surface, 1.2 meters apart, to form a reference trajectory. Using PATH's magnetic guidance system and highly sophisticated signal processing techniques, an onboard computer directs a steering input to an actuator on the steering column, which steers the bus precisely along a desired trajectory, with a tolerance of approximately one centimeter. The bus is also equipped with throttle and brake control systems that allow it to stop within ten centimeters of a designated location. With these capabilities, buses can be automated to dock precisely at bus stops, thus providing easy access and enhancing passenger safety. When combined with a boarding platform that is at the same level as the floor of the bus, precision docking eliminates the need for stepping up into the bus, which can be difficult for elderly or persons with mobility impairments. It also eliminates the need for wheelchair lifts or similar costly devices. The PATH precision-docking bus provides highly reliable and accurate performance. Several LED lights on the dashboard inform the driver about the status and readiness of the bus and the docking systems. The driver can easily select between full and partial automation, and make the transition smoothly.

Because the guidance system eliminates driver variation and driver error in steering, a guided vehicle is capable of precise lane-keeping. The result can be a significant reduction in accidents involving side collisions to transit buses, collisions at bus stops, and passenger injuries during boarding, alighting, bus starting, bus stopping and bus turning. Guided pathways could also reduce scrapes at narrow toll booths and at the bus wash.

PATH's technical staff has considerable experience and expertise in vehicle guidance systems and safety applications. PATH was instrumental in developing and deploying technology for the National AHS Consortium 1997 Demonstration. and many other international technology demonstrations. In recent years, PATH technologies have been deployed on snow removal equipment (snowplow and snow blower) under the sponsorship of Caltrans.

BRT applications are ready for deployment

Many transit agencies across the country are considering the possibilities of implementing BRT technology, and planning their deployment. The technologies are well proven. For deployment, other factors, such as systematic benefit analysis and societal and institutional issues must be evaluated. The California PATH Program, together with the National Bus Rapid Transit Institute (a collaboration between the UC Berkeley Institute of Transportation Studies and the Center for Urban Transportation Research at the University of South Florida) have the expertise to work with public and private agencies to define the requirements as well as deployment paths for specific BRT cases. For more information please phone or email Wei-Bin Zhang, 510-231-9538, e-mail: wbzhang@path.berkeley.edu.