Data processing: generic control systems or specific application – Specific application – apparatus or process – Robot control
Reexamination Certificate
1998-07-13
2001-07-24
Lee, Thomas (Department: 2782)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Robot control
C700S245000, C700S248000, C706S016000, C706S025000
Reexamination Certificate
active
06266577
ABSTRACT:
BACKGROUND OF THE INVENTION
A. Field of the Invention
The present invention relates to robot networks, and, more particularly, to methods and systems for reconfiguring robots of a network.
B. Description of the Related Art
In recent years, robot networks have become more popular and their assigned tasks have become more sophisticated. Typically, robot networks include a plurality of robots that operate according to an on-board control logic. The control logic acts as the “brains” of each robot and defines the action of the robot in response to a sensed input from the environment. For instance, the control logic may define how the robot processes an input data signal or moves in response to a sensed environmental condition. Thus, the performance of the overall robot network is necessarily a function of the accuracy of the control logic.
The article “Issues in Evolutionary Robotics,” Harvey et al., Proceedings of the Second International Conference on Simulation of Adaptive Behavior, (1993), describes a method for reconfiguring the control logic of each robot in a network. Under this approach, each robot is initially downloaded with a different control logic. Thus, the robots of the robot network will each perform the assigned task with a varying degree of success. To improve the performance of those robots that are less successful, a new control logic is determined “off-line.” In particular, genetic programming techniques are used to reconfigure the control logic of the more successful robots to produce a new control logic for the less successful robots. These genetic programming techniques, which include mutation and cross-over techniques, are well known in the art and are used to produce an evolved control logic by reconfiguring the control logic of the more successful robots. After this new control logic is downloaded onto the less successful robots, the robots are then placed back into operation to accomplish an actual task.
A problem with the above approach, however, is that the robots cannot reconfigure their control logic dynamically while performing their assigned task. The new control logic must be determined “off-line.” Nor does this approach allow the robots to reconfigure their control logic using data detected while performing an actual task at hand. In the above approach, the new control logic is determined “off-line” using a predefined set of environmental conditions.
Other robot networks include robots that share information with one another. Thus, data detected by one robot may be shared with the other robots of the network to help those other robots achieve a commonly assigned task. However, these approaches also fail to allow the robots to dynamically reconfigure their control logic. In other words, these networks merely allow robots to share information, but do not allow the robots to change how they process that information.
Therefore, there is a need for a robot network that allows the individual robots to optimally reconfigure their on-board control logic on a real-time basis.
SUMMARY OF THE INVENTION
Systems consistent with the present invention allow robot networks to optimally reconfigure the control logic on-board each robot of the network on a real-time basis.
To achieve these and other advantages, a robot network consistent with the present invention comprises a plurality of robots. Each robot performs an assigned task according to a control logic. To reconfigure the control logic, each robot transmits performance data indicating a fitness level of that robot associated with the assigned task. A subset of the robots is then selected based on the performance data transmitted by each robot. The control logic of the selected subset of robots is then transmitted to other robots of the network. Finally, the control logic of each robot receiving the control logic is reconfigured by producing a new control logic based on the received control logic.
Another aspect of the invention comprises a robot performing an assigned task according to a first control logic. The robot includes a transmitter which transmits performance data indicating a fitness level of the robot associated with the assigned task and a receiver which receives a second control logic determined based on the performance data transmitted by the robot. The robot further includes a processor which optimizes the first control logic of the robot by producing an optimized control logic based on the received second control logic.
Both the foregoing general description and the following Detailed Description are exemplary and are intended to provide further explanation of the invention as claimed.
REFERENCES:
patent: 4788637 (1988-11-01), Tamaru
patent: 4956772 (1990-09-01), Neches
patent: 4998050 (1991-03-01), Nishiyama
patent: 5084826 (1992-01-01), Hariki et al.
patent: 5361023 (1994-11-01), Kim
patent: 5504890 (1996-04-01), Sanford
patent: 5719761 (1998-02-01), Gatti et al.
patent: 5787262 (1998-07-01), Shakib et al.
patent: 5819008 (1998-10-01), Asama et al.
patent: 5825981 (1998-10-01), Matsuda
patent: 5848398 (1998-12-01), Martin et al.
patent: 5907491 (1999-05-01), Canada et al.
patent: 0279558A1 (1988-08-01), None
Koza, John R., Evolution and Co-Evolution of Computer Programs to Control Independently-Acting Agents, European Conference on Artificial Life (ECAL′91), Paris, France, Dec. 1991.
Koza, John R. et al., Automatic Programming of a Time-Optimal Robot Controller and an Analog Electrical Circuit to Implement the Robot Controller by Means fo Genetic Programming, Proceedings of 1997 IEEE International Symposium on Computational Intelligence in Robotics and Automation, Computer Society Press, Los Alamitos, CA, 1997.
Floreano, D. et al., Evolution of Homing Navigation in a Real Mobile Robot, IEEE Transactions on Systems, Man, and Cybernetics-Part B, Cybernetics, vol. 26, No. 3, Jun. 1996.
Floreano, D. et al., Evolution and Learning in Autonomous Robots, in Bio-Inspired Computing Systems, in Mange, D., and Tomassini, M. (eds), PPUR, Lausanne, 1998.
Mondada, F. et al., Evolution and Mobile Autonomous Robotics, in E. Sanchez and M. Tomassini (eds), Towards Evolvable Hardware: The Evolutionary Engineering Approach, vol. 1062 of LNCS, pp. 221-249, Springer-Verlag, 1996.
Nolfi, S et al., How to Evolve Autonomous Robots: Different Approaches in Evolutionary Robotics, Proceedings of the Fourth Workshop on Artificial Life, Boston, Ma, 1994.
Thompson, A., Evolving Electronic Robot Controllers the Exploit Hardware Resources, Submitted to Third European Conference on Artificial Life (ECAL′95).
Inman, Harvey et al., Issues in Evolutionary Robotics, in Meyer, J-A., Roitblat, H.L., and Wilson, S.W. (eds) From Animals to Animats 2, Proceedings of the Second International Conference on Simulation of Adaptive Behavior, pp. 364-373, MIT, 1993.
Reynolds, Craig W., An Evolved, Vision-Based Behavioral Model of Coordinatd Group Motion, in Meyer, J-A., Rotiblat, H.L., and Wilson, S.W. (eds) From Animals to Animats 2, Proceedings of the Second International Conference on Stimulation of Adaptive Behavior, pp. 384-392, MIT, 1993.
Mataric, Maja J., Designing Emergent Behaviors: From Local Interactions to Collective Intelligence, in Meyer, J-A., Roitblat, H.L., and Wilson, S.W. (eds) From Animals to Animats 2, Proceedings of the Second International Conference on Simulation of Adaptive Behavior, pp. 432-441, MIT, 1993.
Floreano, D., Evolutionary Robotics in Behavior Engineering and Artificial Life in Proc. of the 6th Intl. Symposium Evolutionary Robotics ″98 (ER′98), from Intelligent Robotics to Artificial Life, Tokyo, Japan, Apr. 8-9, 1998 (Copy Unavailable).
Jakobi, N. The Minimal Simulation Approach to Evolutionary Robotics, in Proc. of the 6th Intl. Symposium Evolutionary Robotics ′98 (ER′98) from Intelligent Robotics to Artificial Life, Tokyo, Japan, Apr. 8-9, 1998. (Copy Unavailable).
Husbands, P. et al., A Case Study in Evolutionary Robotics in Back, T., Fogel, D., and Michalewicz, Z. (eds) Handbook of Evolutionary Computation, Oxford University Press, 1997. (Copy Unavailable
Montana David J.
Popp Robert L.
Walters Joseph B.
Cao Chun
GTE Internetworking Incorporated
Lee Thomas
Suchyta Leonard Charles
LandOfFree
System for dynamically reconfigure wireless robot network does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for dynamically reconfigure wireless robot network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for dynamically reconfigure wireless robot network will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2512869