Data processing: vehicles – navigation – and relative location – Vehicle control – guidance – operation – or indication – Automatic route guidance vehicle
Patent
1998-06-02
2000-08-22
Chin, Gary
Data processing: vehicles, navigation, and relative location
Vehicle control, guidance, operation, or indication
Automatic route guidance vehicle
701 25, 318587, 180167, G06F16500
Patent
active
061085979
DESCRIPTION:
BRIEF SUMMARY
FIELD OF THE INVENTION
The present invention relates to an autonomous mobile robot system for sensor-based and map-based navigation in pipe networks based on classification of pre-existing landmarks, where said robot system's motion control, sensor information, and landmark classification are allowed to be inaccurate.
BACKGROUND
In the prior art there are known mobile robots for performing in pipe networks. These mobile robots are not autonomous, but leashed by a cable for transmitting energy and control (as for sewer inspection devices) from a remote place outside the pipe network, or they swim passively with the flow of the pipe contents (as in oil or gas pipeline inspection). As far as known to the Applicants, autonomous mobile robots for pipe systems controlling motion actively and having "cut-off the cable" are not known.
For indoor and outdoor environments, autonomous mobile robot systems are known (for example so-called floor robots. I. Nourbakhsh, R. Powers, and S. Birchfield. DERVISH: An office-navigating robot. AI Magazine, 16(2):53-60, 1995.). These known robots are equipped with sensors and maps for navigating using landmarks in the environment as reference points. For many of these approaches, artificial landmarks (beacons) have to be installed in the environment only for the purpose of supporting navigation. Those known robots that make use of natural landmarks often suffer from uncertainty in the landmark classification. If landmarks can be beyond the sensory horizon and drift is possible in the robot motion, navigation is impeded by overlooking landmarks. Moreover, all the known autonomous mobile robots suffer from inaccuracy of motion like drift, slip, overshooting.
In the literature, models and algorithms are known (like so-called POMDPs which is an acronym for Partially Observable Markov Decision Processes for Artificial Intelligence) that take sensor and motion inaccuracy into account. POMDPs are described for example in L. Kaelbling, M. L. Littman, and A. R. Cassandra. Partially observable markov decision processes for artificial intelligence. In I. Wachsmuth, C. -R. Rollinger, and W. Brauer, editors, KI-95: Advances in Artificial Intelligence, pages 1-17. Springer Verlag (LNAI vol.981), 1995, and R. Simmons and S. Koenig. Probabilistic robot navigation in partially observable environments. In Proc. IJCAI-95, pages 1080-1087, 1995. These algorithms and models work the tore poorly in following a given path, the more likely it is to overlook landmarks. Due to the friction of the cables, cable-driven devices have a very limited range of operation in "natural" pipe networks. On the other side, map-based and sensor-based navigation of autonomous mobile robots is impeded by overlooking landmarks.
It is the object of the present invention to provide a map-based and sensor-based mobile robot system in a network of pipes wherein the robot system is autonomous and has no cable connection to the outside and, wherein the mobile robot system is capable of using "natural" landmarks inherently in the pipe network.
SUMMARY OF THE INVENTION
According to the invention, this object is solved in an autonomous mobile robot system for sensor-based and map-based navigation in pipe networks based on classification of pre-existing landmarks, where said robot system's motion control, sensor information, and landmark classification are allowed to be inaccurate, said system comprising of a starting point to a destination point in terms of said topological map, according to said path plan; the control of said motion is allowed to be inaccurate, said vehicle along its axis of motion and the position of the vehicle within the pipe at the pipe bottom; notified with certainty of the presence of any landmark that the vehicle passes while navigating through the pipe network, landmarks in terms of given landmark classes; said classification is allowed to be inaccurate, class of the landmarks to be encountered while following said path plan through the pipe network, plausible position of said vehicle based on the e
REFERENCES:
patent: 4500970 (1985-02-01), Daemmer
patent: 4829442 (1989-05-01), Kadanoff et al.
patent: 5001635 (1991-03-01), Yasutomi et al.
patent: 5032775 (1991-07-01), Mizuno et al.
patent: 5204814 (1993-04-01), Noonan et al.
patent: 5363305 (1994-11-01), Cox et al.
patent: 5525883 (1996-06-01), Avitzour
patent: 5758298 (1998-05-01), Guldner
patent: 5911767 (1999-06-01), Garibotto et al.
Proc. IJCAI-95, 1995, USA pp. 1080-1087, XP000197391.
AT Magazine, vol. 16(2) pp. 53-60.
8095 IEEE Transations on Industry Applications 30 (1994) Sep./Oct. pp. 1321-25.
IROS '94 Proceedings of the IEEE/RSJ/GI International Conference on Intelligent Robots and Systems; pp. 298-305.
Hertzberg Joachim
Kirchner Frank
Chin Gary
GMD-Forschungszentrum Informationstechnik GmbH
LandOfFree
Autonomous mobile robot system for sensor-based and map-based na does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Autonomous mobile robot system for sensor-based and map-based na, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Autonomous mobile robot system for sensor-based and map-based na will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-592647