Unmanned Ground Vehicles
Intelligent Vehicle Systems
Contact Information
Ryan D. Lamm
Assistant Director
Intelligent Vehicle Systems & Networks
+1 (210) 522-5350
rlamm@swri.org
SwRI has developed a fully autonomous vehicle that can coordinate maneuvers with other vehicles and navigate obstacles in an urban environment.
Recent dramatic advances led by defense science programs, such as the Defense Advanced Research Projects Agency (DARPA) Grand and Urban Challenges, and car and truck Original Equipment Manufacturer (OEM) industry advances in active vehicle safety systems provide evidence that the field of advanced vehicular situational awareness and computer control/automation of vehicles is entering a rapid growth phase. In addition, the U.S. Department of Defense (DOD) and Congress plan for one-third of all ground combat vehicles to be autonomous by 2015.
In response to these industry goals, Southwest Research Institute (SwRI) initiated a $5 million internal research and development program in late 2006 to investigate and develop sensor/computing/mobile technologies to augment vehicle platforms and provide autonomous vehicle capabilities to improve safety in urban and traffic environments. Through the Mobile Autonomous Robotics Technology Initiative (MARTI®), SwRI is developing technology associated with the autonomous control of cars, trucks, and tractors.
Enabling Technologies
The technical breadth of SwRI allowed for the assembly of a multidisciplinary team of scientists and engineers with extensive capabilities and experience working in fields related to intelligent vehicle systems. On MARTI, this team applied its respective domain knowledge to include the current state of the art in:
- Perception
- Intelligence
- Command and control
- Communications
- Platforms
- Safety
This knowledge is representative of how SwRI typically forms a multidisciplinary project team to address specific technical challenges.
The MARTI Program
SwRI began the MARTI program by conducting an extensive survey of the the current state of the art. This included an analysis of existing technology, current relevant research, and the technology gaps that existed in industry. SwRI engineers identified cutting edge hardware (to include some prototype sensors) and developed modular and scalable complex software algorithms that resulted in a fully autonomous ground vehicle capable of negotiating a complex urban environment.
Within 12 months, the sensors were integrated; preliminary object identification, classification, and tracking algorithms were developed; and the vehicle was under computer control. The vehicle was capable of operating autonomously around the SwRI test track facility at various speeds. Within 18 months, MARTI was negotiating intersections in mixed traffic obeying traffic laws. The system was demonstrated at the ITS World Congress (New York City) in November 2008, alongside Stanford University/VW.
Additional information about the autonomous vehicle demonstration is available.
SwRI and INRIA also demonstrated a jointly developed cooperative vehicle-to-vehicle pedestrian safety system at the ITS World Congress in November 2008 in New York City. A patent is pending for this technology.
SwRI and INRIA demonstrated a jointly developed cooperative
vehicle-to-vehicle pedestrian safety system at
the ITS World Congress in November 2008 in New York City.
Additional information about the cooperative vehicle-to-vehicle pedestrian safety system is available.
Expected outcomes from this program include follow-on R&D in the following areas:
- Robust driver-assist and active-safety systems
- Algorithms supporting autonomous operation in less-constrained environments
- Cooperative vehicle-highway systems
- Strategic coordination of objectives among multiple autonomous ground vehicles
Subsequent to the commercial application urban environment demonstration in New York City, MARTI has been augmented to support convoy logistics and support collaborative multi-vehicle mission behaviors in military applications. A demonstration was held at the Robotics Rodeo at Ft. Hood, Texas, in September 2009. MARTI was demonstrated in a convoy environment highlighting how an Unmanned Ground Vehicle (UGV) can be utilized to enhance convoy operations.
SwRI engineers developed MARTI® for commercial
applications, including convoy operations, using
off-the-shelf components.
In addition to the automotive and military applications, the MARTI technology has been applied to solve complex challenges in mining, demining, construction, agricultural, and other industries.
Heavy-duty prototype system. Click image for a larger view.
SwRI’s UGV was developed as part of a $5 million internal research and development program to investigate and develop sensor, computing and mobile technologies. These could be utilized to augment a commercial off-the-shelf vehicle platform to provide autonomous vehicle capabilities that can operate and perform specific tasks and improve safety and mobility in urban and trafficked environments. The integrated autonomy hardware controls the SwRI’s unmanned autonomous vehicle. Click the image for a larger view.
The MARTI Vehicle
The base vehicle, a commercially available Sport Utility Vehicle (SUV), was mechanically and electrically modified to allow for installation of the various:
- Actuators
- Sensors
- Computer platforms
- Support software
- Dedicated Short Range Communications (DSRC) prototype radios, which are identical to the ones used on the USDOT Connected Vehicle Program
- Command and control real-time control software
- Data logger
- System health monitor
This platform is being used to benchmark various hardware and software intelligent vehicle systems and is available to support advanced research and development of autonomous and active safety systems for a variety of customers.
Technology Applications
MARTI is the largest internal research and development program undertaken by SwRI. The MARTI technology was developed with commercial challenges in mind and was augmented and proven to address military applications.
Through MARTI, SwRI has developed approaches and algorithms that can facilitate:
- Active vehicle safety systems including collision avoidanc
- Autonomous vehicle navigation including complex vehicle behaviors and maneuvers such as intersection negotiation
- Advanced situational awareness: identification, classification, and tracking of objects in the environment
- Sensor fusion of various commercially available and prototype sensors
- Operation in GPS-limited environments
- Special applications such as mission completion in adverse environments
- Vehicle-to-vehicle and vehicle-to-infrastructure communication
- Cooperative systems to include interaction with unattended ground sensors and cooperative sensor sharing
- Convoy operations
- Unattended ground targets
Related Terminology
intelligent transportation systems • ITS • automation and data systems • southwest safe transport initiative • autonomous control of cars • military vehicles • DARPA • active safety devices • vehicle autonomy
