B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In Proc. of 5th Int. Symposium on Distributed Autonomous Robotic Systems (DARS), 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....maximum gain from the environment. This gain is the key point in the learning phase because it implicitly introduces the desired behaviour of the robots. In this research, the CMOMMT [20] application has been used as a test bed. This test bed is being widely used in distributed robotics research [28, 3] and allows us to make several comparisons with previous works [21] The next section describes the basis of the VQQL technique and its use in reinforcement learning problems, as well as a brief description of the Gen eralized Lloyd Algorithm and Q Learning. Section 3 describes the CMOMMT ....

Barry Brian Werger and Maja Matarid. Broadcast of local eligibility for multi-target observation. In L.E. Parker, G. Bekey, and J. Barhem, edi- tors, Distributed Autonomous Robotic Systems, volume 4, pages 347-356. Springer, 2000.

....the vacancy chain algorithm sensitive to different levels of ability in a group of heterogeneous robots as well as to the effects of the underlying group dynamics such as interference and synergy. 1 Introduction Existing task allocation (TA) algorithms [Parker, 1997; Botelho and Alami, 1999; Werger and Matari c, 2000; Gerkey and Matari c, 2002] are not sensitive to the complex effects of group dynamics, such as interference and synergy. For a cooperative task such as transportation or foraging, the average task completion time depends on the number of robots that are allocated to the same task. Allocating a ....

B. B. Werger and M. J. Matari c. Broadcast of local eligibility for multi-target observation. In Proceedings of the Fifth International Symposium on Distributed Autonomous Robotic Systems (DARS'00), pages 347--356, Knoxville, TN, October 4-6 2000. Springer.

....in all cases, and that the best task allocation strategy changes as a function of the noise in the system. This result is significant, and shows the need for further investigation of task allocation strategies. 1 Introduction There has been significant prior research in multi robot coordination [1, 12, 2, 14, 3, 15, 7]. We view this problem as an instance of dynamic task allocation. Presently, a general theory of task allocation for multi robot domains remains elusive. This paper empirically derives general guidelines for selecting task allocation strategies for multirobot systems. The guidelines are ....

....to perform the task; 2) an auctioning mechanism decides which robot gets the task; 3) the winning robot s controller performs a sequence of actions to execute the task. In [14] each robot s ability to perform a task is mapped to a scalar quantity, which is used to assign tasks to robots. In [15], a local eligibility mechanism is described as the robots perceived ability to perform a task. We use the approach from [6] to construct a general formulation for the multi robot coordination problem. In this formulation, a bidding function determines each robot s ability to perform a task ....

[Article contains additional citation context not shown here]

B. Werger and M. Mataric. Broadcast of local eligibility for multi-target observation. In Proceedings, 5th International Symposium on Distributed Autonomous Robotic Systems (DARS), Knoxville, TN, Oct 4-6, pages 347--356, Oct. 2000.

....is Alliance [9] a behavior based soft ware architecture for heterogeneous multi robot coop eration. It has a fault tolerance mechanism that al lows the robots to detect failures in their teammates and adapt their behaviors to complete the task. An other behavior based approach is presented in [13], in which robots broadcast messages with their eligibility in order to coordinate their actions in a multitarget observation task. Although some approaches propose coordination methods without the use of ex plicit communication (for example [7] the develop ment of cheaper and more reliable ....

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 6. Springer Verlag, 2000.

....the tracking network, and they can also adapt to the movement of targets or the dynamic changes in an environment by re positioning themselves in response. We introduce a scalable, cooperative tracking system that consists of multiple mobile robots and multiple stationary sensors. Prior research [7, 8] on cooperative tracking problem has focused on allocating multiple robots to multiple targets in a bounded environment with no obstacles. This work is supported in part by DARPA grant DABT63 99 10015 and NSF grants ANI 9979457 and ANI 0082498 These systems were not applied or scaled for ....

....to track multiple moving targets. There is relatively little research on cooperative multitarget tracking using multiple robots. 7] utilized the ALLIANCE architecture to achieve target assignment; it used implicit communication (acquiescence and impatience levels) for cooperation. In contrast [8] used explicit communication (inhibition signal over the network) for target al..location. 10] described a Variable Structure Interacting Multiple Model (VS IMM) estimator combined with an assignment algorithm for tracking multiple ground targets using multiple stationary sensors. As different ....

Barry B. Werger and Maja J. Mataric, "Broadcast of local eligibility for multi-target observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000.

.... and they can adapt to the movement of the targets (e.g. follow targets to occluded areas) The robot based target tracking problem (CMOMMT: Cooperative Multirobot Observation of Multiple Moving Targets) has been formally de ned in [1] and has received recent attention in the robotics community[2, 3]. The CMOMMT problem is de ned as follows. Given a bounded, enclosed region S, a team of m robots R, a set of n targets O(t) and a binary variable In(o j (t) S) de ned to be true when target o j (t) is located within region S at time t, and m n matrix A(x) is de ned where a ij (t) 8 : 1 ....

....However, it was assumed that the observation sensors had a perfect eld of view and a known global coordinate system. Experiments were performed in a bounded, enclosed spatial region, and an indoor global positioning system was utilized as a substitute for vision or rangesensor based tracking. In [3], an approach to a similar problem using the BLE (Broadcast of Local Eligibility) technique was presented which used a real video camera to track moving objects, and one way communication for explicit role assignment. However, the environment in [3] was very simple, and movements of targets were ....

[Article contains additional citation context not shown here]

Barry B. Werger and Maja J. Mataric, \Broadcast of local eligibility for multi-target observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000.

No context found.

B. B. Werger and M. J. Mataric, Distributed Autonomous Robotic Systems 4. Springer-Verlag, 2000, ch. Broadcast of Local Eligibility for Multi-Target Observation, p. 347356.

No context found.

B. Werger and M. Mataric. Broadcast of local eligibility for multi-target observation. In Proceedings, of DARS., pages 347-356, Oct. 2000.

No context found.

Barry Brian Werger and Maja J Matari c. Broadcast of Local Eligibility for Multi-Target Observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, pages 347--356. Springer-Verlag, 2000.

No context found.

Barry B. Werger and Maja J Mataric. Broadcast of local eligibility for multi-target observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, Proceedings of the 5th International Symposium on Distributed, Autonomous Robotic Systems (DARS'00), pages 347--356, Knoxville, Tennessee, October 4-6 2000. Springer.

No context found.

B. B. Werger and M. J. Mataric , "Broadcast of local eligibility for multitarget observation," in Distributed Autonomous Robotic Systems 4,L. E. Parker, G. Bekey, and J. Barhen, Eds. New York: Springer-Verlag, 2000, pp. 347--356.

No context found.

Barry Brian Werger and Maja J Matari c. Broadcast of Local Eligibility for Multi-Target Observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, pages 347--356. Springer-Verlag, 2000.

No context found.

Barry Brian Werger and Maja J Matari c. Broadcast of Local Eligibility for Multi-Target Observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, pages 347--356. Springer-Verlag, 2000.

....and in that case, both pushing robots could directly perceive the goal, somewhat reducing the need for cooperation. A great deal of work has also been reported on multiagent coordination systems, though the agents themselves are seldom physically embodied. Notable exceptions are BLE [13], and the ALLIANCE architecture [10] both of which have been applied to a multi target tracking task with groups of robots. ALLIANCE was also applied to a box pushing task [9] but the system runs open loop, whereas we have closed the control loop through the use of our watcher, with the ....

Barry Brian Werger and Maja J Mataric. Broadcast of Local Eligibility for Multi-Target Observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, pages 347--356. SpringerVerlag, 2000.

....that using softmax functions makes the vacancy chain algorithm sensitive to different levels of ability in a group of heterogeneous robots as well as to the effects of the underlying group dynamics such as interference and synergy. 1 Introduction Existing task allocation (TA) algorithms [12, 4, 15, 8] are not sensitive to the complex effects of group dynamics, such as interference and synergy. For a cooperative task such as transportation or foraging, the average task completion time may depend on the number of robots that are allocated to the same task. Allocating a robot to a task may have ....

B. B. Werger and M. J. Matari c. Broadcast of local eligibility for multi-target observation. In Proceedings of the Fifth International Symposium on Distributed Autonomous Robotic Systems (DARS'00), pages 347--356, Knoxville, TN, October 4-6 2000. Springer.

....us c. edu applets emergency The 6 Related work There has been significant prior research in multi robot collaboration [1, 9, 2] The ALLIANCE architecture [10] presents a robust, multi robot, task allocation system. An opportunistic approach based on mutual inhibition is presented in [12]. An approach based on commitment is presented in [5] where a task allocation strategy using a market based auction system commits the robots to their tasks until success or failure. Previous work on sound based servoing in robotics includes [7] in which a single robot was able to servo to ward ....

B. Werger and M. Matarid. Broadcast of local eligibility for multi-target observation. In Proceedings, of DARS., pages 347-356, Oct. 2000.

.... the case with many of the approaches described below) Specifically, assuming that all available tasks are currently underway, each engaged robot Name Computational Communication Task Requirements iteration Requirements iteration Consideration ALLIANCE [21] m A h reassignment BLE [23] m A h reassignment M [3] h no reassignment MURDOCH [13] 7 auctioneer no reassignment First price auctions [10, 24] 7 bidder Dynamic role assignment [6] 7 Table 1: Summary of MRTA architectures broadcasts only a heartbeat ....

....algorithm: for each available task, compare its own utility to that of every other robot and select the shortest task for which it is most capable (thus tasks are implicitly prioritized by length) This algorithm can be executed in time per iteration. Broadcast of Local Eligibility (BLE) [23] is another behavior based approach to MRTA, with fixed priority tasks. For each task, each robot has a corresponding behavior that is capable of executing the task, as well as estimating the robot s utility for the task. Utilities are computed in a task specific manner as a function of relevant ....

Barry Brian Werger and Maja J Mataric. Broadcast of Local Eligibility for Multi-Target Observation. In Lynne E. Parker, George Bekey, and Jacob Barhen, editors, Distributed Autonomous Robotic Systems 4, pages 347--356. SpringerVerlag, 2000.

....The question of whether a controller (distributed or centralized) which has knowledge of the number of individual robots and their capabilities, and history of targets observed, could generate better task distributions remains open for future study. Details of this work can be found in [24, 23]. 5 Auction Based Task Allocation 5.1 The Problem Domain: BoxPushing The experimental problem domain we address here is cooperative planar manipulation, specifically moving a box, large relative to the size of the robots, from some initial location to some observable (to at least one robot) goal ....

Barry Brian Werger and Maja J Matari'c. Broadcast of local eligibility for multi-target observation. In Proceedings, 5th International Symposium on Distributed Autonomous Robotic Systems (DARS), pages 347--356, Knoxville, TN, 2000.

....and differences among the three approaches, as well as to address the problem at a more general level. The first approach we describe is opportunistic (robots grab a task if they think they are eligible to perform it) and uses broadcast of local eligibility and mutual inhibition among robots [22]. The second approach, in contrast involves commitment to the selected task [8] and task allocation is based on a market based auction system. The third approach [15] studies the tradeoff between opportunism and commitment relative to cooperative vs. mutually exclusive individuals in a multirobot ....

....of Mars. There has been significant prior research in multirobot collaboration, including [2, 11, 4] The ALLIANCE architecture [17] presents a robust, multirobot, task allocation system. An opportunistic approach based on mutual inhibition and broadcast of local eligibility is presented in [22]. The approach is applied to a target tracking problem studied previously in [18, 19] An auction based approach based on commitment is presented in [8] where a task allocation strategy using a market based auction system commits the robots to their tasks until success or failure. The latter two ....

B. Werger and M. Mataric. Broadcast of local eligibility for multi-target observation. In Proceedings, of DARS., Oct. 2000.

....present in Section 4.1. 4 A Scalable Example: Multi Target Observation We now present an example of how effective multi robot strategies can be implemented with a small number of connections and extremely simple component behaviors using Ayllu. This example is a system that performs the W CMOMMT [20,15] cooperative multi robot observation of multiple moving targets task. In W CMOMMT, the goal is for a group of robots to keep a number of moving targets under observation (that is, in the field of view of the robot and within a meter or so) Each target has a priority, so that when there are more ....

....to separate targets (rather than redundantly observing some targets while leaving others uncovered) while adapting to changing numbers of robots and targets and robot failures, we use a technique called the Broadcast of Local Eligibility. Our BLE approach to W CMOMMT is thoroughly detailed in [20]. Ayllu: Distributed Port Arbitrated Behavior Based Control 7 a) Output Inhibit Best Local Robot 1 Behavior n Output Inhibit Best Local Robot Behavior n Output Inhibit Best Local Robot Behavior n b) Search Search Observe Observe Observe Observe Search Observe Observe ....

[Article contains additional citation context not shown here]

Barry Brian Werger and Maja J Mataric. Broadcast of local eligibility for multi-target observation. submitted to DARS2000.

No context found.

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In Proc. of 5th Int. Symposium on Distributed Autonomous Robotic Systems (DARS), 2000.

No context found.

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In DARS00, pages 347--356, 2000.

No context found.

B. B. Werger and M. J. Mataric, "Broadcast of local eligibility for multi-target observation," in Proc. of 5th Int. Symposium on Distributed Autonomous Robotic Systems (DARS), 2000.

No context found.

B. B. Werger and M. J. Mataric, "Broadcast of local eligibility for multi-target observation," in DARS00, 2000, pp. 347--356.

No context found.

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In Proceedings of Distributed Autonomous Robotic Systems, pages 347-356, 2000.

No context found.

B. B. Werger and M. J. Mataric, "Broadcast of local eligibility for multi-target observation," in Proc. of 5th Int. Symposium on Distributed Autonomous Robotic Systems (DARS), 2000.

No context found.

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In DARS00, pages 347--356, 2000.

No context found.

B. B. Werger and M. Mataric, "Broadcast of local eligibility for multitarget observation," in Distributed Autonomous Robotic Systems 4, L. E. Parker, G. Bekey, and J. Barhen, Eds. Springer-Verlag, 2000, pp. 347-- 356.

No context found.

Barry B. Werger and Maja J. Mataric, "Broadcast of local eligibility for multi-target observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000.

No context found.

Barry B. Werger and Maja J. Mataric, "Broadcast of local eligibility for multi-target observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000.

No context found.

B. B. Werger and M. J. Matari c, "Broadcast of local eligibility for multitarget observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000, pp. 347--356.

No context found.

Werger, Barry B., and Maja J Matari c, "Broadcast of local eligibility for multi-target observation", Distributed Autonomous Robotic Systems 4, Proceedings of the 5th International Symposium on Distributed, Autonomous Robotic Systems (DARS'00) (Knoxville, Tennessee, October 4-6, 2000) (L. E. Parker, G. Bekey, and J. Barhen eds.), Springer, 347-- 356.

No context found.

B. B. Werger and M. J. Mataric, "Broadcast of local eligibility for multitarget observation," in Proceedings of Distributed Autonomous Robotic Systems, 2000.

No context found.

B. B. Werger and M. J. Mataric. Broadcast of local eligibility for multi-target observation. In Proc. of DARS, 2000.

No context found.

Werger, Barry B., and Maja J Matari c, "Broadcast of local eligibility for multi-target observation", Distributed Autonomous Robotic Systems 4, Proceedings of the 5th International Symposium on Distributed, Autonomous Robotic Systems (DARS'00) (Knoxville, Tennessee, October 4-6, 2000) (L. E. Parker, G. Bekey, and J. Barhen eds.), Springer, 347-- 356.

No context found.

Barry Brian Werger and Maja Mataric. Broadcast of local eligibility for multi-target observation. In L.E. Parker, G. Bekey, and J. Barhem, editors, Distributed Autonomous Robotic Systems, volume 4, pages 347-356. Springer, 2000.

No context found.

B.B. Werger and M.J. Mataric. (2000). Broadcast of local eligibility for multi- target observation. In Proceedings of DARS 4, 347-356. 14

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC