Adaptive Robotics and Intelligent Systems
INL’s Robotics and Intelligent Systems division specializes in using existing platforms to build smarter robots that can adapt on the fly to changing environments, communication interruptions, mission requirements and other challenges in the field. The robotics program dates back several decades and was originally created to provide remote handling capabilities for the laboratory in its role as a nuclear testing station. In the early 1990s, the group branched into formal robotics, focusing on unmanned autonomous systems. As a result, INL adaptive robotic researchers developed the Robotic Intelligence Kernel, a cutting-edge hardware, software and sensor technology that enhances a robot’s capacity to interpret and “understand” its surroundings. The low-cost onboard control architecture gives robots exceptional levels of autonomy and intelligence. The RIK won a R&D 100 Award and a Stoel Rives Idaho Innovation Award in 2006. INL researchers are now concentrating on developing “dynamic path planning” robots that gather information in the field, process it using an experience knowledge base and then decide the best route to their destination.
- Unmanned Ground Vehicles -Robotic and Intelligent Systems scientists design unmanned ground vehicles from ATVs, Segways and other machines that can perform a variety of tasks. One area in which they have excelled is creating robots to detect landmines. INL researchers, working with the U.S. Army Maneuver Support Center, Carnegie Mellon University, Night Vision Laboratories and the Space and Naval Warfare Command, have developed portable, low-cost autonomous robots that identify and mark buried landmines. The results of a rigorous 10-day experiment showed that RIK running on a Carnegie Mellon University robot detected and accurately marked, both physically and digitally, 130 out of 135 buried mines. There was not a single false detection during the test. The robot was able to navigate obstacles while conducting area searches and the de-mining of roads and dismounted lanes. It swept a 50-meter lane and accomplished those tasks in a few minutes compared to the 25 minutes or it takes a trained soldier – and there was no human put in harm’s way.
- Unmanned Aerial Vehicles -INL adaptive robotic researchers specialize in designing low-cost unmanned aerial vehicles for surveillance, hyperspectral imaging, air quality testing and other uses that do not require humans to remotely pilot them. One of the primary UAVs being used by the INL team is manufactured by Arcturus. The T-16 is a rugged UAV suitable for applications where reliability and extended persistence are required. Typical endurance ranges from 12 to 24 hours depending on payload weight and fuel load. The T-16 has a generous fuselage providing 729 cu. in. of internal payload space. 12 lbs of fuel can be carried in the T-16's wings, freeing up the entire fuselage for payload mounting if required. The T-16 is integrated with Cloud Cap Technology's GPS based Piccolo II autopilot, providing autonomous launch, flight, and landing. The UAVs have been used for rapid, accurate, and safe collection of ultra high-resolution geographically referenced aerial imagery. Utilizing a 16-mega pixel camera, ultra high-resolution images are captured from an altitude of 800 to 1000’ and in near real-time wirelessly streamed to a ground-based operating interface where each image is automatically mosaiced into a single geo-referenced map. The system was designed in direct response to the Air Force’s need for a rapid airfield damage assessment immediately following a strategic attack on an airfield or upon initial evaluation of an enemy airfield. INL researchers are now working on an unmanned vehicle system that combines multiple assets including UAVs, UGVs, and unattended ground sensors. The system gleans data from aerial and ground surveillance and subsequently generates mission plans for each of the assets rather than depending on a pre-programmed route. This "dynamic path planning" allows robots to have greater autonomy and to adapt to unforeseen obstacles in the field and mission changes.
- Remote Systems and Automation -INL’s Robotics and Intelligent Systems division has long developed products that allow hazardous materials to be handled remotely. The program’s goal since its inception is to minimize human exposure to potentially dangerous situations, chemicals and materials. It focuses on creating robotic and automated systems to do the work instead of depending on humans to “suit up” and handle the materials remotely. It was instrumental in developing a system to remotely open 55-gallon drums of mixed waste, automatically sort the contents into acceptable and non-compliant waste streams, and repackage the waste for shipment. Recently they just completed development and demonstration of a system to remotely seal and leak test waste containers for disposal in the Yucca Mountain waste repository.