Author name cluster

Mazda Ahmadi

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

AAMAS Conference 2007 Conference Paper

IFSA: Incremental Feature-Set Augmentation for Reinforcement Learning Tasks

Mazda Ahmadi
Matthew E. Taylor
Peter Stone

Reinforcement learning is a popular and successful framework for many agent-related problems because only limited environmental feedback is necessary for learning. While many algorithms exist to learn effective policies in such problems, learning is often used to solve real world problems, which typically have large state spaces, and therefore suffer from the "curse of dimensionality. " One effectivemethod for speeding-up reinforcement learning algorithms is to leverage expert knowledge. In this paper, we propose a method for dynamically augmenting the agent's feature set in order to speed up value-function-based reinforcement learning. The domain expert divides the feature set into a series of subsets such that a novel problem concept can be learned from each successive subset. Domain knowledge is also used to order the feature subsets in order of their importance for learning. Our algorithm uses the ordered feature subsets to learn tasks significantly faster than if the entire feature set is used from the start. Incremental Feature-Set Augmentation (IFSA) is fully implemented and tested in three different domains: Gridworld, Blackjack and RoboCup Soccer Keepaway. All experiments show that IFSA can significantly speed up learning and motivates the applicability of this novel RL method.

PDF

ICRA Conference 2006 Conference Paper

A Multi-robot System for Continuous Area Sweeping Tasks

Mazda Ahmadi
Peter Stone 0001

As mobile robots become increasingly autonomous over extended periods of time, opportunities arise for their use on repetitive tasks. We define and implement behaviors for a class of such tasks that we call continuous area sweeping tasks. A continuous area sweeping task is one in which a group of robots must repeatedly visit all points in a fixed area, possibly with nonuniform frequency, as specified by a task-dependent cost function. Examples of problems that need continuous area sweeping are trash removal in a large building and routine surveillance. In our previous work we have introduced a single-robot approach to this problem. In this paper, we extend that approach to multi-robot scenarios. The focus of this paper is adaptive and decentralized task assignment in continuous area sweeping problems, with the aim of ensuring stability in environments with dynamic factors, such as robot malfunctions or the addition of new robots to the team. Our proposed negotiation-based approach is fully implemented and tested both in simulation and on physical robots

Details

AAAI Conference 2006 Short Paper

Biconnected Structure for Multi-Robot Systems

Mazda Ahmadi

Many applications of distributed autonomous robotic systems can benefit from, or even may require, the team of robots staying within communication connectivity. For example, consider the problem of multirobot surveillance, in which a team of robots must collaboratively patrol a given area. If any two robots can directly communicate at all times, the robots can coordinate for efficient behavior. This condition holds trivially in environments that are smaller than the robots' communication range. However in larger environments, the robots must actively maintain physical locations such that any two robots can communicate — possibly through a series of other robots. Otherwise, the robots may lose track of each others’ activities and become miscoordinated. Furthermore, since robots are relatively unreliable and/or may need to change tasks (for example if a robot is suddenly called by a human user to perform some other task), in a stable multirobot surveillance system, if one of the robots leaves or crashes, the rest should still be able to communicate. Some examples of other tasks that could benefit from any pair of robots being able to communicate with each other, are multirobot exploration, search and rescue, and cleaning robots.

PDF Details

AAAI Conference 2006 Conference Paper

Keeping in Touch: Maintaining Biconnected Structure by Homogeneous Robots

Mazda Ahmadi

For many distributed autonomous robotic systems, it is important to maintain communication connectivity among the robots. That is, each robot must be able to communicate with each other robot, perhaps through a series of other robots. Ideally, this property should be robust to the removal of any single robot from the system. In (Ahmadi & Stone 2006a) we define a property of a team’s communication graph that ensures this property, called biconnectivity. In that paper, a distributed algorithm to check if a team of robots is biconnected and its correctness proof are also presented. In this paper we provide distributed algorithms to add and remove robots to/from a multi-robot team while maintaining the biconnected property. These two algorithms are implemented and tested in the Player/Stage simulator.

PDF Details