Robust Robotics
Abstract
(U) The Robust Robotics program is developing advanced robotic technologies that will enable autonomous (unmanned) mobile platforms to perceive, understand, and model their environment; navigate through complex, irregular, and hazardous terrain; manipulate objects without human control or intervention; make intelligent decisions corresponding to previously programmed goals; and interact cooperatively with other autonomous and manned vehicles. These capabilities will enable robotic vehicles to support warfighters in diverse environments including urban, ground, air, space, and underwater. A key objective is intelligent control of mobile manipulators to independently perform subtasks over a broad range of domains of interest to the warfighter, thereby reducing operator workload, time on target, training time, bandwidth, and hardware complexity. Another key objective is robust navigation and locomotion even in the absence of GPS, since this underlies the ability to move through the difficult and unpredictable terrain of theater operations, which may include highly irregular and mountainous areas, partially-destroyed roads, rubble-filled urban terrain, and other vehicles and personnel. Robust Robotics is also developing techniques for robots to perform in dynamic environments by improving robotic vision and scene understanding. This includes the capability to predict the future location and even the intent of moving objects in order that robots can handle both movement and clutter simultaneously and plan a collision-free course through the environment. Future autonomous systems must also achieve a much higher autonomy level when performing complex tasks, and so Robust Robotics is developing techniques that will enable robotic agents to achieve effective levels of autonomous reasoning whether humans are present or not. Future robotic agents must also be able to effectively perform when they are part of a team and assume semi-independent roles across a variety of activities. This will be achieved by developing robotic systems that can accept and understand instructions to define new activities and their variants from human controllers.
Document Details
- Document Type
- Accomplishment
- Publication Date
- Oct 01, 2011
- Source ID
- 51028ebc764559255f3e52defbe4070e