Southwest Safe Transport Initiative (SSTI), 10-R9648

Printer Friendly Version

Principal Investigator
Steven W. Dellenback, Ph.D., PMP

Inclusive Dates:  07/24/06 – 01/31/09

Background - Technology needed to facilitate the autonomous control of cars, trucks, tractors, and military vehicles has made dramatic advances in the last few years led by defense science programs such as the Defense Advanced Research Projects Agency (DARPA) Grand Challenge, DARPA Urban Challenge, and OEM car and truck industry advances in active safety systems and associated sensors. Furthermore, there is evidence that the field of autonomous ground vehicle technology is entering a rapid growth phase. For instance, the U.S. Department of Defense, along with a U.S. Congressional objective, has stated that one-third of ground combat vehicles will be unmanned by 2015. At the same time, the U.S. Department of Transportation is investing heavily in the development of communications technology needed to link the intelligent highway with the intelligent vehicle through its Vehicle Infrastructure Integration (VII) program. The continued convergence of these industries will allow for significant advancements to be achieved in intelligent vehicles systems as the component technologies are transferred across industries and applied in new and exciting ways.

The Southwest Safe Transport Initiative (SSTI) was initiated to improve safety in urban traffic environments. SSTI is charged with developing new sensor, computing, and mobile technologies to augment vehicle platforms and provide autonomous vehicle capabilities.

Approach - SSTI is fusing the latest technology from multiple industries to meet the challenges associated with autonomous control of cars, trucks, tractors, and military vehicles. Through the incorporation of technologies and design methodologies from multiple industries such as unmanned aerial systems, intelligent transportation systems, cognitive multi-agent systems, machine vision, engineering dynamics, hardware and software in-the-loop simulation, large-scale multi-function robotics, and safety and reliability systems, SwRI is developing a full-scale autonomous ground vehicle platform to be used for advanced engineering of intelligent vehicle systems and applications development.

The base vehicle, a commercially available SUV, was mechanically and electrically modified to allow for installation of perception sensors; computer platforms and support software; communications, command and control system; data logger; and system health monitor. Substantial effort is being put into the identification of existing hardware systems, which would be scalable to allow for augmentation, modular enough to allow components to be removed from the system while maintaining system functionality, flexible enough to allow parallel development of different components, and reliable through the use of robust and redundant systems.

The hardware selection was based on a balance of the capabilities, size, ease of use, and cost of the various existing systems. The perception system consists of sensors and software necessary to identify the environment, obstacles, terrain and internal state of the vehicle. A sophisticated Global Position System/Inertial Navigation System (GPS/INS) was installed on the vehicle to allow for determination of the vehicle’s location. This GPS/INS also aids with precise location information that supports route planning and vehicle control. Intelligence is implemented using hardware and software for situation awareness, knowledge representation and planning and execution. Another interface to the intelligence will be the communications hardware. Dedicated Short Range Communication (DSRC) radios are being used for the vehicle-to-roadside communications and vehicle-to-vehicle communications to facilitate the development of advanced cooperative, multi-agent, multi-vehicle systems.

The command and control hardware interprets path information from intelligence and implements the necessary commands to move the vehicle such that the error between the planned path and the actual path is minimized. The hardware is based on a computer platform running under a true, real-time operating system. The system sends the necessary commands and receives feedback from the drive-by-wire system actuators, which control the vehicle transmission, brake, accelerator, and steering.

Accomplishments - The project is currently about one-third of the way through the development cycle. SwRI engineers have installed various prototype sensors targeted for future automotive applications and high performance blade computing hardware utilized for computationally intensive algorithms, developed extensive vehicle dynamic models to optimize control algorithms, established rudimentary software control of the vehicle, and optimized near real-time control algorithms to allow the SSTI vehicle to follow a dense set of waypoints at up to 50 km per hour around the SwRI 1.16-mile oval test track.

Currently path planning, object recognition and other sensor input are being integrated into the vehicle. Once the SSTI vehicle is fully autonomous, SwRI intends to utilize this platform for advanced research in vehicle behaviors, intelligence and knowledge representation, cooperative vehicle maneuvers and interactions, and in the development of advanced situational awareness. SwRI has already seen benefits from applying technology from the ITS industry to this technical challenge and believes strongly that the ITS industry can benefit from the technology development needed to make vehicles less human-dependent.

2007 Program Home