Tsuyoshi Ueyama

Concept of cellular robotic system (CEBOT) and basic strategies for its realization

This paper presents the concept of Cellular Robotic System (CEBOT), which is one of the self-organizing robotic systems, and describes some basic strategies for its realization. CEBOT is the distributed intelligent system. This system is a Dynamically Reconfigurable Robotic System (DRRS) consisting of a large number of autonomous robotic units called cells. At first, for the determination of CEBOT structure, we describe the reconfiguration methodology for mobile/robotic manipulator systems, which is based on geometric calculation. Because of the important issue that CEBOT can automatically connect and communicate with each other, then we show both the experimental results in automatic connection and separation by CEBOT MarkII designed as a protoype of CEBOT, and the communication experiments in CEBOT MarkII. Finally, we propose the optimal knowledge allocation method based on the information flow of the communication among cells, since the optimal knowledge allocation is considered as one of the most important issues for the distributed intelligent system. In this paper, we describe the ability of CEBOT and show some experimental results.

Self-organization of cellular robots using random walk with simple rules

An evaluation of a self-organization method of cellular robots using random walk with simple rules is presented. The approach for the self-organization of cellular robots is based on the behavior of primitive lives such as bacterium. The cellular robots or cellular robotic system called CEBOT consists of a large number of autonomous robotic units named cells. Simple rules are adopted to each cell for organization of the cellular robots autonomously and distributively. In the process of self-organization of cellular robots, each cell must communicate with another cell, and negotiate with them, because the organization of cellular robots cause competition of cells as resources among cells. It is assumed that each cell has genetic information as a structure map itself. Simulation results show the effectiveness of the self-organization of the cellular robots depending on several parameters, the balance of selfishness and cooperation and the number of cells with the genetic information.<<ETX>>

Hierarchical control architecture for Cellular Robotic System-simulations and experiments

Describes the hierarchical control architecture of real mobile robots for Cellular Robotic System Mark-V (CEBOT Mark-V). A parallel processing control system has been adopted, and by adjusting the role of parallel processes standing for the states of independent primitive behaviors, the change of system organization is realized to adapt the redefinition of plural tasks and the variation of environments. The authors propose a method for decision making of a mobile robots behavior through integrating multiple behavioral processes. The authors define two relation matrices denoting the relationship among the processes: the priority matrix and the interest relation matrix. The matrices are used to adjust the outputs of behavioral processes and optimize the behavior of mobile robots. To obtain the most suitable priority matrix, the authors introduce a learning-adapting algorithm. The results of simulation and experiment with a real CEBOT Mark-V showed the effectiveness of the proposed matrices and learning-adapting algorithm. On the other hand, instead of simply selecting processes for decision making of the robots behavior, the integration of multiple processes based on the proposed matrices enhanced the control robustness of robot system.

Smooth path planning by using visibility graph-like method

To achieve smooth motion of car-like robots, it is necessary to generate paths that satisfy the following conditions: maximum curvature, maximum curvature derivative, and curvature continuity. Another requirement is that human operators can manipulate the robots with ease. In this paper, a path expression methodology consisting of line segments, circular arcs and clothoid arcs is presented. In addition, a method of global path generation with a visibility graph is proposed. To establish this method, the following steps are proposed: (a) the arrangement of subgoals (middle points) and (b) the construction of the graph for path generation. By using the proposed method, the paths were shortened 14% on average.

Design of an artificial mark to determine 3D pose by monocular vision

The design for an artificial mark attached to small objects is presented in this paper. Three conditions are required for the design of the marks: (1) short calculation time, (2) easy attachment, and (3) individual calculation for each item. The marks are obtained by using three different extraction methods. The experiments determine the best mark. The mark using the method that extract colors and calculated the gravity has good results. We analyze the errors of the mark that has best experiment. The appropriateness of the experimental results is then confirmed. We have the experiment that the manipulator handles the objects with the best mark. The experiment is successful. Therefore, the usability of the mark is verified.

Control strategy for a network of cellular robots-determination of a master cell for cellular robotic network based on a potential energy

A control strategy for a network of cellular robots (CEBOT) is proposed. The strategy is based on the selection of a master cell and assumes centralized control of the CEBOT network. The selection is made using a network energy, in which an energy function is calculated based on the information flow in the CEBOT network and the state space of the communication transition probability. The authors describe both a method of mastering, which implies the determination of the master cell of the network based on the information flow between cells, and a method of mastering by a local energy sensitive estimation as a method of finding the master cell for the CEBOT network. It is assumed that the network has a binary tree structure for control and communication. The master cell is defined as a window for interaction between users and the CEBOT. Simulation results on the mastering are shown.<<ETX>>

Coordinated motion control of a robot arm and a positioning table with arrangement of multiple goals

The minimum-time motion coordination is an important subject in robotics. In this study, the arrangement of several goals, which is treated as a traveling salesman problem (TSP), is incorporated to this subject. Although TSP has been studied in most previous works; but, solving a TSP that takes into account collision occurrences has not received much attention. This instance arises when a robot arm has to plan a sequence of reaching goals with other moving objects and/or other robot arms. If goals are also moving, then the problem becomes more complex since the end configuration of robot arm is undefined when reaching a goal. In this study, in particular, a 6-DOF robot arm has to reach several goals found in an object while a 1-axis positioning table simultaneously positions the object; thereby changing the goal locations and collision occurrences are inevitable. For the purpose of this study, the TSP is solved effectively with motion coordination and collision avoidance. The collision-free configurations of a robot arm when reaching goals are solved through motion coordination. Collision is avoided by exploiting the redundancy of the system. The above-mentioned solution is verified through a simulation utilizing an object with various numbers of goals and their positions, and is proven effective.

Practical Point-to-Point Multiple-Goal Task Realization in a Robot Arm with a Rotating Table

This study presents a multiple-goal task realization in a system composed of a 6-d.o.f. robot arm and a one-axis rotating table. The problem is complex due to the existence of multiple goals and the kinematic redundancy of the system. We propose a design approach integrating the base placement, task sequencing and motion coordination methods. We show that this approach reduces the task completion time of the robot arm; the motion planning is realized through straight-line paths in the configuration space despite collision occurrences. Furthermore, we introduce a hybrid graph-search method combining the greedy nearest-neighbor method and the Dijkstra method to solve the motion coordination of the robot arm and the table. We show the effectiveness of the design approach and the search method through a time-constrained simulation-based optimization.

Compact design of work cell with robot arm and positioning table under a task completion time constraint

A work cell is generally designed to achieve a high throughput and its size is typically viewed as contingent to component sizes. In this paper, we aim to design a compact work cell (spatial requirement) and to minimize its task completion time (temporal requirement) to a value set as a constraint. By doing so, a work cell occupies a minimal space and achieves its desired throughput. The work cell size is evaluated based on the size and the swept volume of components. This evaluation is important since a robot arm can have a very large swept volume depending on a given task. To satisfy the spatial and temporal requirements, we propose the integration of the base placement optimization, goal rearrangement, and motion coordination between the robot arm and the positioning table. Furthermore, we introduce two motion coordination schemes based on the spatial and temporal requirements. We showed the effectiveness of the proposed method through simulations.

Direct teaching and error recovery method for assembly task based on a transition process of a constraint condition

A method for simplification of teaching in robotic assembly is presented. Generally, an assembly task is a process in which a constraint condition between two workpieces changes. There has been much research on teaching an assembly task to a robot system by a model-based approach. However, it takes a lot of time to revise information of models that depend on the task environment when every new task is taught to a robot in these model-based approaches. We propose a method to simplify teaching by taking an approach where a human teaches a task to a robot without giving models of the task environment and task processes. Additionally, we also propose a method to realize robotic assembly that is robust to errors by executing a task as a transition process of a constraint condition. Experimental results show the effectiveness of our method.