Munehiro Takimoto

Higher-Order Mobile Agents for Controlling Intelligent Robots

This paper presents a framework for controlling intelligent robots connected by the Internet. This framework provides novel methods to control coordinated systems using higher-order mobile agents. Higher-order mobile agents are hierarchically structured agents that can contain other mobile agents. By using higher-order mobile agents, intelligent robots in action can acquire new functionalities dynamically as well as exchange their roles with other colleague robots. The higher-order property of the mobile agents enables them to be organized hierarchically and dynamically. In addition to these advantages, higher-order mobile agents require minimum communication. They only need connection to be established when they perform migration.

Saving Energy Consumption of Multi-robots Using Higher-Order Mobile Agents

This paper presents a framework for controlling intelligent robots connected by communication networks. This framework provides novel methods to control coordinated systems using higher-order mobile agents. Higher-order mobile agents are hierarchically structured agents that can contain other mobile agents. The combination of the higher-order mobile agent and mobile multi-robots open a new horizon of efficient use of mobile robot resources. Instead of physical movement of multi-robots, mobile software agents can migrate from one robot to another so that they can find the most suitably equipped and/or the most suitably located robots to perform their task. Thus the framework presented here provides efficient use of robot resources. In this paper, we focus on the energy saving. We have demonstrated the efficiency by numerical experiments.

Suppressing the Total Costs of Executing Tasks Using Mobile Agents

This paper presents a framework for controlling multiple robots connected by communication networks. Instead of making multiple robots pursue several tasks simultaneously, the framework makes mobile software agents migrate from one robot to another to perform the tasks. Since mobile software agents can migrate to arbitrary robots by wireless communication networks, they can find the most suitably equipped and/or the most suitably located robots to perform their task. In the previous papers, we have showed that this manner of controlling multiple robots can decrease the number of required robot resources, and therefore can suppress energy consumption in aggregation. In this paper, we focus on the efficiency aspect of the manner of controlling multiple robots. Our controlling manner for multiple robot is expected to achieve further contribution for suppressing the total costs of executing tasks. We demonstrate the effectiveness of our manner of controlling multiple robot by numerical experiments.

Design of a Multi-Robot System Using Mobile Agents with Ant Colony Clustering

This paper presents a framework for controlling mobile multiple robots connected by communication networks. This framework provides novel methods to control coordinated systems using mobile agents. Instead of physical movement of multiple robots, mobile software agents can migrate from one robot to another so that they can minimize energy consumption in aggregation. In some applications, it is desirable that multiple robots draw themselves together automatically. In order to avoid excessive energy consumption, we employ mobile software agents to locate robots scattered in a field, and make them autonomously determine their moving behaviors by using a clustering algorithm based on the Ant Colony Optimization (ACO). ACO is the swarm intelligencebased method that exploits artificial stigmergy for the solution of combinatorial optimization problems. Preliminary experiments have provided a favorable result. In this paper, we focus on the implementation

Searching targets using mobile agents in a large scale multi-robot environment

This paper presents a framework for controlling multiple robots connected by communication networks. Instead of making multiple robots pursue several tasks simultaneously, the framework makes mobile software agents migrate from one robot to another to perform the tasks. Since mobile software agents can migrate to arbitrary robots by wireless communication networks, they can find the most suitably equipped and/or the most suitably located robots to perform their task. In the previous papers, we have shown that this manner of controlling multiple robots can decrease the number of required robot resources in the three-robot case. The three robots demonstrated that they could suppress the energy consumption through a mobile agent based approach. In this paper, we pursue a case of large scale multiple robots.We have implemented a simulator for a large number of robots based on our framework, and have demonstrated the efficiency and the scalability. We report the observations we found in the numerical experiments.

Ant colony clustering using mobile agents as ants and pheromone

This paper presents a new approach for controlling mobile multiple robots connected by communication networks. The control mechanism is based on a specific Ant Colony Clustering (ACC) algorithm. In traditional ACC, an ant convey an object, but in our approach, the ant implemented as a mobile software agent controls the robot corresponding to an objects, so that the object moves to the direction required by the ant agent. At this time, the process where an ant searches an object corresponds to some migrations of the ant agent, which are much more efficient than physically searching. Also, the ACC uses a pheromone for making a cluster grown and stabilized. However, it is difficult to implement such a pheromone as an physical entity, because it can diffuse, mutually intensify its strength, and restrict its effect in its scope. In our approach, the pheromone is implemented as a mobile software agent as well as an ant. The mobile software agents can migrate from one robot to another, so that they can diffuse over robots within their scopes. In addition, since they have their strength as vector values, they can represent mutually intensifying as synthesis of vectors. We have been developing elemental techniques for controlling multiple robots using mobile software agents, and showed effectiveness of applying them to the previous ACC approach which requires a host for centrally controlling robots. The new ACC approach decentralizes it, and makes a robot system free from special devices for checking locations.

A serialization algorithm for mobile robots using mobile agents with distributed ant colony clustering

This paper presents effective extensions of our previously proposed algorithm for controlling multiple robots. The robots are connected by communication networks, and the controlling algorithm is based on a specific Ant Colony Clustering (ACC) algorithm. In traditional ACC, imaginary ants convey imaginary objects for classifying them based on some similarities, but in our algorithm, we implemented the ants as actual mobile software agents that control the mobile robots which are corresponding to objects. The ant agent as a software agent guides the mobile robot (object) to which direction it should move. In the previous approach, we implemented not only the ant but also the pheromone as mobile software agents to assemble the mobile robots with as little energy consumption as possible. In our new approach, we take advantage of the pheromone agents not only to assemble the robots but also to serialize them. The serializing property is desirable for particular applications such as gathering carts in airports. We achieve the property by allowing each ant agent to alternatively receive a pheromone agent. We have built a simulator based on our algorithm, and conducted numerical experiments to demonstrate the feasibility of our approach. The experimental results show the effectiveness of our algorithm.

Design of an intelligent cart system for common airports

This paper presents a framework for an intelligent cart system designed to be used in common airports. This framework provides novel methods to control carts using mobile software agents. In airport terminals, it is desirable that carts draw themselves together automatically after being used so that manual collection becomes less laborious. In order to avoid excessive energy consumption by the carts, we employ mobile software agents and RFID to locate carts scattered in a field and cause them to autonomously determine their moving behavior using clustering based on the ant colony optimization (ACO) algorithm.

Distributed Evacuation Route Planning Using Mobile Agents

This paper proposes a distributed multi-agent framework for discovering and optimizing evacuation routes on demand. Our framework assumes mobile ad hoc networks (MANETs) composed of smartphones with geo-location capabilities. On the network, heterogeneous mobile agents cooperatively insert knowledge about crowd in our mass evacuation framework. They are relying exclusively on crowd sourcing; therefore our framework is layout independent and adaptable for any situation. The mobile agents take advantage of ant colony optimization (ACO) in order to collect such knowledge. Once users reach safe areas, they distribute agents to inform the directions of the locations of the safe areas. On the other hand, evacuating users distribute agents to search safe areas, based on guidance given by the agents from the safe areas. Once each searching agent reaches the safe area, it traces its path backwardly collecting geographical information of intermediate nodes for composing an evacuation route. During the backward travel, agents lay down pheromone as they migrate back based on the ACO algorithm, strengthening quasi-optimal physical routes, and hence guiding succeeding agents. A characteristic of pheromone in this family of algorithms is that it lessens during run-time, keeping the information about successful escape routes current, as is essential in an evacuation scenario. We have implemented a simulator based on our framework in order to show the effectiveness of our technique. We discuss the behaviors of our system with various settings on the simulator for real world implementation in the near future.

Controlling biped walking robots using genetic algorithms in mobile agent environment

The design and implementation of a control program for biped walking robots using the genetic algorithms (GA) are presented. The most difficult problem with biped walking robots is that they have too many possible gaits. Generally it is impossible to find the optimal gait for a given route. In order to control biped walking robots, we have employed GA to determine the gaits of the robots. It is known that keeping the zero moment point (ZMP) in certain area is necessary for stable movement of a biped walking robot deriving ZMP is not theoretically difficult; it is just a matter of solving kinetic equations. The problem is that, for a certain series of ZMPs, the robot can have too many gaits and too hard to find the optimal one. We are using a control program using GA to produce approximately optimal gaits.