The RoverBug algorithm plans the motion of a robotic vehicle (rover) equipped with a CCD (charge-coupled device) camera or comparable sensor that has a limited field of view. The algorithm produces locally optimal (shortest-distance) paths across unbounded, previously unknown terrain; utilizes gaze control to minimize the amount of sensing; and avoids unnecessary robot motion. Because this algorithm is based on a local and simple mathematical model of terrain, it does not entail the bookkeeping of a global model (with all of the attendant issues of registration of local maps with global ones and with each other), and does not require the large memory that would be required by a global model. The algorithm is amenable to varying levels of autonomy, ranging from single- subpath execution under tight operator guidance to completely autonomous traverses to distant goals. In addition to being useful for exploration of distant planets via rovers, the algorithm could be applied in such terrestrial scenarios as cleanup of environmental hazards or military surveillance. A version of the algorithm has been implemented on NASA’s Rocky 7 prototype microrover.
This work was done by Sharon Laubach of Caltech for NASA’s Jet Propulsion Laboratory.
This software is available for commercial licensing. Please contact Don Hart of the California Institute of Technology at (818) 393- 3425. Refer to NPO-30241.
This Brief includes a Technical Support Package (TSP).
Algorithm Plans Motion of Robot with Limited Field of View
(reference NPO-30241) is currently available for download from the TSP library.
Don't have an account?
Overview
The document outlines the development and capabilities of the RoverBug algorithm, a motion planning system designed for robotic vehicles, particularly rovers, equipped with sensors like CCD cameras that have a limited field of view. This algorithm is significant for its ability to navigate unbounded, previously unknown terrains while producing locally optimal paths that minimize travel distance. It employs gaze control to reduce the amount of sensing required, thereby avoiding unnecessary movements of the robot.
One of the key advantages of the RoverBug algorithm is its reliance on a local and simple mathematical model of the terrain, which eliminates the need for complex global models that require extensive memory and bookkeeping. This feature is particularly beneficial given the constraints faced by planetary rovers, such as limited computational capacity, power, and memory. The algorithm is designed to operate efficiently within these constraints, making it suitable for various applications, including planetary exploration, environmental hazard cleanup, and military surveillance.
The document also highlights the algorithm's implementation on NASA's Rocky 7 prototype microrover, showcasing its practical application in simulated Martian terrain. The RoverBug algorithm is described as complete and correct, ensuring that it can reliably achieve specified targets with minimal ground intervention. This capability is crucial for missions with limited lifetimes and operational windows, as it allows for more efficient use of available resources.
Additionally, the document discusses the broader context of motion planning for planetary rovers, emphasizing the need for planners that can operate under severe constraints while still achieving effective navigation. The RoverBug algorithm addresses these needs by producing paths that are not only optimal but also require minimal memory and processing power.
In summary, the RoverBug algorithm represents a significant advancement in robotic motion planning, particularly for applications in challenging environments like those found on other planets. Its efficient use of resources, combined with its ability to navigate complex terrains autonomously, positions it as a valuable tool for future exploration missions. The work was conducted by Sharon Laubach at Caltech for NASA's Jet Propulsion Laboratory, and the software is available for commercial licensing.
