Social insects demonstrate adaptive behaviour for a given colony size. Remarkably, most species do this even without visual information in a dark environment. However, how they achieve this is yet unknown. Based on individual trait expression, an agent-based simulation was used to identify an explicit mechanism for understanding colony size dependent behaviour. Through repeated physical contact between the queen and individual workers, individual colony members monitor the physiological states of others, reflecting such contact information in their physiology and behaviour. Feedback between the sensing of physiological states and the corresponding behaviour patterns leads to self-organisation with colonies shifting according to their size. We showed (1) the queen can exhibit adaptive behaviour patterns for the increase in colony size while density per space remains unchanged, and (2) such physical constraints can underlie the adaptive switching of colony stages from successful patrol behaviour to unsuccessful patrol behaviour, which leads to constant ovary development (production of reproductive castes). The feedback loops embedded in the queen between the perception of internal states of the workers and behavioural patterns can explain the adaptive behaviour as a function of colony size.

This paper proposes a collective construction method through interaction between simple robots and intelligent blocks that have a rule set and a communication function with neighboring blocks. In our method, the structure is formed by growing chain of blocks. The growth direction is determined by the rule set and a counter value passed between the blocks. The robots load or unload the block based on a simple algorithm and a local signal from the blocks. Because of the simplicity of each robot's behavior, the structure is locally unstable: the blocks can be attached or detached randomly from the structure even if they have already formed a part of the structure. In this paper, we first explain the mechanism of our proposal. Next we show some fundamental characteristics obtained by computer simulation and small scale experiment by real robot and blocks. We introduce small modification to enhance the adaptability of the system mainly focusing on a reinforcement function of the wall against the external stress.

We propose a novel algorithm that allows a small team of robots to fully explore an unknown environment, even in the face of extreme communication restrictions. We verify, through proofs and experiments, that the algorithm will achieve full exploration.

In this paper, author present a system which named Triple Furuta pendulum. This system has three pendulums which placed at direct drive motor. Author aim at a realizetion of a stabilization of a Triple Furuta pendulum. For the stabilizing control, author derived the equation of motion by the Eular-Lagrange method and a state equation. Author show structural evaluation based on the condition number of controllability matrix and design of present system. Since there are three pendulums, each link would be affected by disturbance. And therefore a simulation of stabilizing control by sliding mode control with high robustness.

This paper reports collective behaviors of multirobot system with simple dynamics and interactions. The model for collective motion is described by fundamental kinetics and the dynamics of the heading which each element has as a degree of freedom. Firstly, we shows the system based on this model realizes various types of behavior according to the set of parameters in the model, which resembles to the behavior of livings such as fish, birds and small insects. Next, we refer to the behavior of the modified model improving the anisotropy of the interaction force. Especially, we discuss the characteristic of linear formation which is obtained by modifying the term of optimal distance between neighbors. Performance of the model is examined by both computer simulation and robot experiments

Division of labor performed by social insects, such as ants and bees, is one of the most advanced functions that they have evolved. Although individual insects have the same inheritance, they are divided into certain groups in which they play specialized roles. For effective performance, the groups exhibit proportional regulation of population, which implies regulating their population to maintain their proportion; further the proportion of each group is regulated against external disturbances without aid from a special individual. In a multi-robot system, the division of labor is one of the most attractive and useful propositions. This paper addresses the proportion regulation of population in a homogenous multi-robot system, which is inspired by the task allocation practiced by social insects. We propose an extended probability model for proportion regulation, and confirm its performance by numerical simulation. We also experimentally validate its performance using a simple multi-robot system.

In multi-robot system, communication is indispensable for effective cooperative working. In this system, direct communication by physical methods such as light, sound, radio wave is quite general. But in biological system, especially in the insect world, not only the physical but also the chemical communication methods can be observed. As the chemical methods have some unique properties, it is challenging to apply such a method to the cooperative multi-robot system. Unfortunately, to treat real chemical materials for the robots is not easy for now because of some technical difficulties. In this paper, we propose virtual pheromone system in which chemical signals are simulated with the graphics projected on the floor, and in which the robots decide their action depending on the color information of the graphics. We examined the performance of this system through the foraging task, which is one of the most popular tasks for multi-robot system and is generally observed in ant societies.

Today, most effective method to estimate the function of unknown protein is to retrieve similar known sequences. There are some effective techniques for sequence retrieval to estimate protein functions. We propose a method to express the amino acid sequence by finite state automaton. At first, we introduce an accumulated amino acid residue score (ARS). Next, we introduce a finite state automaton (FSA) to describe the alignment of amino acid residues. Combining ARS and FSA, and we can calculate the similarity of known protein sequence and unknown one. We applied this method to a protein family and examined the efficiency of this method.

Proteins have complicated spatial structure and have chemical and physical functions that originate from the structure. Today no method is available to predict the function accurately from the DNA/Amino acid sequence. Instead, there are some approaches to estimate the functions approximately based on a similarity retrieval of sequences. In this paper, we propose two types of methods for amino acid sequence retrieval by an evolutionary dictionary. One is based on homology retrieval. Introduction of the compression by evolutionary dictionary technique enables us to describe the text data as an n-dimensional vector using n dictionaries which are generated by compressing n typical texts, and it also enables us to classify them based on their sequential similarity. The other is based on motif retrieval. As there are some common arrangements in functionally similar amino acid sequences, we can make a "dictionary" which is specific to the group. In this method, we introduce a genetic algorithm and refine the dictionary. Effectiveness of our proposal is examined using real genome data.

We introduce the concept of the ill-configured object (ICO). An ICO is a geometrical object having a stable (unique) name but varying configurations (shape, size, components, and component layout). In addition, we introduce the concept of the neighbour set representation (NSR) of an object, and show that the NSR is well-fitted to design an algorithm to detect ICOs in images. Two applications of the proposed theory to aerial image analysis are then presented.

Research of multirobot systems is active in these days. The most remarkable characteristic of multirobot system is that the robots work cooperatively and achieve the task which a single robot cannot do. This characteristic is important and it is essential to investigate an effect of number of robots in the system. There are some tasks which art! suitable for multirobot system and we chose foraging task. A robot used in this paper has a simple interaction method with a light signal. In this paper, we discuss their behavior in a field that has one feeding point at first. Their behavior is analyzed by a robot simulation and a mathematical model. Next, we discuss a performance of the system in a clockface arranged foraging field, in which some localized foods are located equidistant from the home. Here, we report the group shows an ordered behavior depending on the number of robots and the strength of the interaction.

We propose a novel method of self-organizing formation. It is assumed that elements are not connected to each other and they can move in continuous space. The objective is to arrange elements in a certain spatial pattern like a crystal, and to make the outline of the group in the desired shape. For this purpose, we propose a method by using virtual springs among the elements. In this algorithm, an element generates virtual springs between the neighbor element based on information of how many other elements exist in the neighborhood with a certain radius. Although the elements interact locally, only by virtual springs, and they do not have global information at all, they form a shape much larger than the sensory radius. By a simulation study, we confirmed convergence to a target shape from a random state in very high probability. This kind of algorithm gives a new principle of self-organizing formation, and its simplicity will be useful for the design of self-assembling nano machines in future.

We propose the PRDC (Pattern Representation based on Data Compression) scheme for media data analysis. PRDC is composed of two parts: an encoder that translates input data into text and a set of text compressors to generate a compression-ratio vector (CV). The CV is used as a feature of the input data. By preparing a set of media-specific encoders, PRDC becomes widely applicable. Analysis tasks - both categorization (class formation) and recognition (classification) - can be realized using CVs. After a mathematical discussion on the realizability of PRDC, the wide applicability of this scheme is demonstrated through the automatic categorization and/or recognition of music, voices, genomes, handwritten sketches and color images.

Protein has a complicated spatial structure and has chemical and physical functions that originate from the structure. It is important to predict the structure and functions of the protein from a DNA sequence or an amino acid sequence from a viewpoint of biology, medical science, protein engineering and so on. But at present no method is available to predict them accurately from the sequence. Instead, there are some approaches to estimate the functions approximately based on a similarity retrieval of sequences. We propose a new method for similarity retrieval of amino acid sequence based on the concept of homology retrieval using data compression. Introduction of the compression by dictionary technique enables us to describe the text data as an n-dimensional vector using n dictionaries which are generated by compressing n typical texts, and it also enables us to classify them based on their similarity. To classify the data clearly, it is effective to use ideal character strings as dictionaries. In this paper, we introduce genetic algorithm for dictionary generation and classify the amino acid sequences. Effectiveness of our proposal is examined using real genome data.

A method for building time series prediction model using genetic programming is proposed. The construction of prediction models consists of two stages. The first stage composes the most appropriate functional form with genetic programming. The second stage fixes optimal parameters involved in the composite function with a backpropagation-like algorithm. The second stage can be recognized as a local search, which is a powerful tool to accelerate the evolving speed of GP and GA. The method is applied to typical time series and some real world prediction problems. Results of computer generated chaotic time series were compared to those of neural network based and autoregressive predictions. The superiority of the proposed method is demonstrated in the results of these experiments.