In our laboratory, we have succeeded in acquiring forward actions to various robot systems using Reinforcement Learning. We have also succeeded in acquiring a giant swing motion as dynamic task by devising its rewards. Then, the purpose of this study is to investigate the effect of probabilistic behavior which gives to the giant swing motion acquisition. Although the giant swing robot has a continuous dynamic motion such as its angle and angler velocity, its state of the motion must be divided into 216 states in order to apply the reinforcement learning. For this reason, this robot shows probabilistic behaviors. This study investigated the relationship between the performance of learning knowledge and rotation count of the learning process. Consequently, it became clear that the performance of the knowledge is deteriorated when the ratio of knowledge use of the learning process is large.

抄録全体を表示

Passive Dynamic Walking is a technique that makes a robot walk naturally like a human adjusting to skews and slight slopes without active control or energy. It is energy efficient, but to set the initial condition for robots to walk stably is difficult. So when a Passive Dynamic Walking model is affected by a slip, it is not considered to walk stably. This paper analyzes how the gait stability of Passive Dynamic Walking robots is affected by a slip. The results of this study provide some ideas about how to apply such technique in a variety of environments or conditions.

抄録全体を表示

The purpose of this research is to develop an automatic design method for hardware and software of machine. In this paper, the genetic algorithm is applied as the design method for shape and behavior of a virtual mobile modular robot. Mobile robots consist of the same parts, and the purpose of the design is to increase a moving distance. Designed mobile robots are simulated in a three-dimensional virtual physics space, and evaluated by moving distance. As a result, an appropriate shape and behavior for virtual robots is acquired.

抄録全体を表示

In this study, we try to acquire autonomous behaviors of the Frog model that is available in both environments on ground and under water. The behaviors of the artificial organism are controlled by embedded ANN controllers in a virtual 3D physical environment. The ANN controllers are optimized by using particle swarm optimization method for getting appropriate behaviors to accomplish given tasks. A frog model is defined as a typical organism which can adapt in on ground environment and under water environment. Obtained results of computer simulations show a usefulness of our approach.

抄録全体を表示

In this paper, we develop an associative memorization architecture of the musical features from time sequential data of the music audio signals. This associative memorization architecture is constructed by using of Restricted Boltzmann Machine (RBM) and Conditional RBM (CRBM). Challenging purpose of our research is development of the new composition system that automatically creates a new music based on some existing music. In this composition system, characteristics or common musical features of existing music are learned and extracted, then a new different musical melody having such musical features is created. This paper illustrates the important former part of the whole system.

抄録全体を表示

The Job-shop Scheduling Problem is a classical, but yet modern problem. Because it belongs to the NP-hard problem and it is almost impossible to find out the exact solution, many heuristic methods have been proposed. Recently, Genetic Algorithm (GA), Local Clustering Organization (LCO) and so on have been proposed as a powerful tool to solve the problem. The optimal schedule becomes very important from a viewpoint of cost reduction in the manufacturing industry. This study optimizes the job-shop scheduling with multiple purpose functions by using evolutional computation. Numerical experiments verify that the evolutional computation obtains (quasi-) optimal schedule with multiple purpose functions.

抄録全体を表示

In recent years, various robots have been developed as technology has advanced, operating not only in factories but also in places like homes and cities. However, many of those robots are fundamentally set up to repeat the same operation, and have difficulty with environmental changes and unexpected situations. Reinforcement learning is a method of solving this problem. Reinforcement learning is one of the machine learning methods. The learning is a technique of finding an optimal behavior pattern with respect to the target depending on the circumstances and environment by repeating trial and error. In this study, we applied reinforcement learning to a robot modeled on a four-legged animal and examined whether the robot could attain running behavior.

抄録全体を表示

This paper describes a convergence estimation method and learning termination method for reinforcement learning in dynamic environment. In recent years, multi-robot systems utilizing reinforcement learning have been developing in real-world situations. However, conventional learning methods take a considerable amount of time to reach convergence. Furthermore, conventional learning processes are often inefficient because robot continues reinforcement learning even if learning converges. In response to this problem, we propose a Knowledge Co-creation Framework (KCF) for multi-robot systems, whose efficient implementation requires an autonomous convergence estimation method and learning termination method for reinforcement learning. Therefore,on basis of the assumption that learning curves exhibit fractality, we propose the convergence estimation method and the learning termination method utilizing a fractal dimensional analysis. Furthermore, we confirmed that the proposed method determines the learning convergence and terminates the reinforcement learning by conducting a computer simulation.

抄録全体を表示

Twist of seaweeds is one of the physical problems that disturb the growth of the seaweeds in the marine biomass cultivation. This study aims to evaluate the twist of seaweeds in a quantitative way. There is no computational model to simulate dynamics of the twist of seaweeds. In this study, we create the physical model of the seaweeds using physics simulation and the states between seaweed individuals are characterized with feature vectors, which is composed the physical, geometrical, and time factors. The measure of twist can be defined as the number of features and classifies a state between individuals into "twist state" or "non-twist state".

抄録全体を表示

Job-shop Scheduling Problem (JSP) is one of the representative combinatorial optimization problems where the objective is to find an ideal assignment of resources when multiple tasks must be processed simultaneously. Many heuristic methods for the JSP has been proposed. Among them, dispatching rules can provide a feasible schedule within a relatively short time. However, the obtained schedule is usually insufficient and further improvement by other heuristics is needed. Sub-problem is defined by assignments of resources within an arbitrary selected time window and rule-based reconstruction is performed by solving the sub-problem with a dispatching rule. An approprate set of time window and dispatching rule is obtained by genetic algorithm. This rule-based reconstruction procedure is combined with tabu search algorithm. The experiments are shows that the proposed algorithm performs better than an existing heuristic method when the number of tasks is much larger than the number of resources.

抄録全体を表示

In recent years, a multi-agent robot system (MARS) utilizing reinforcement learning and transfer learning has been deployed in real-world situations. Autonomous agents of MARS obtain behavior autonomously by multi-agent reinforcement learning method, and a transfer learning method enables to reuse the knowledge of other robot's behavior such as cooperative behavior. Those methodologies, however, have not been fully discussed systematically. Hence, we propose the knowledge co-creation framework leveraging the transfer learning and a cloud computing. Until now, we developed a Hierarchical Transfer Learning (HTL) as core technology of a knowledge co-creation and indicated effectiveness of the HTL in a dynamic multi agent environment. However, an effectiveness of our proposed HTL depends on transfer rate and approximation accuracy of knowledge generated by artificial neural network. In this paper, we evaluate the effectiveness of the HTL in new artificial neural network settings.

抄録全体を表示

The job-shop scheduling problem (JSP) is one of the combinatorial optimizations and one of the most difficult order decision problems. Many approximate algorithms, which are heuristic methods or metaheuristics, are studied to solve JSP, because JSP is categorized into NP-hard problems. The application of the dispatching rules is known as a practical heuristic for JSP or many other scheduling problems. This study proposes an extended algorithm for JSP based on the genetic algorithm and a novel local search method using the dynamic dispatching rules. The extended algorithm searches for effective dispatching rules along with schedules according to the genetic algorithm. Also the schedules are partially modified by the rules in every generation. In order to verify the effectiveness of the proposed algorithm, this paper shows the result of numerical experiments.

抄録全体を表示

This paper deals with a kinematics resolution framework for robots controller to perform multi target positioning. The framework consists of metaphysical subsystems, corresponding to joint and link, and their connection rules. The localized forward and inverse kinematics calculations are embedded in the subsystems and kinematics models are composed of a set of subsystems assembled by their connection rules. The framework is applied to case studies of the forward and the inverse kinematics problems which are multi target positioning of five or more d.o.f. robot manipulator.

抄録全体を表示

Environment exploration of dynamic unknown environments is an attractive research issue. In this paper, we propose an evolutionary based method to deal with this issue. We evolved a neural controller that uses the laser data and information of unexplored environment to generate the robot best action. The map of the environment is generated by the robot in a form of a binary matrix. The explored are is used as fitness function of the genetic algorithm. The experimental results show a good performance of the proposed method for exploration of unknown dynamic environments.

抄録全体を表示

Reinforcement Learning is a machine learning method to acquire a series of actions that maximizes a cumulative reward. However, it is difficult to optimize interaction between human and robot in a daily living space because there is no definite evaluation standard about undesirable actions. In this study, we propose a novel learning model using a successive reward and punishment based on human subjective evaluation. In this method, we developed human can restrain undesirable actions by giving punishment evaluation. We developed a dog-like robot to verify the proposed method and demonstrated its performance through the experiment.

抄録全体を表示

Sphere-shaped robots are expected to be useful, in that they can move around in home or office environments without getting stuck, even in a dusty room with a lot of obstacles. In order to provides both omni-directional mobility and a large inner space, the authors have previously proposed a new driving scheme employing an omni-directional three-wheel internal unit. Then, a prototype robot was fabricated and constant speed movements were achieved experimentally. However, due to the phere shape, it is observed that the robot shakes largely when it stops running. Therefore in this presentation, reduction of swing movements after stopping is considered. First, a scheme is proposed in which the rolling and pitching angular velocities of the internal unit are measured using a gyro sensor and feedback to velocity reference signals for omni-wheels. Second, the scheme is implemented on the prototype robot. Finally, Experiments using the prototype are described which confirms the validity of the proposed scheme.

抄録全体を表示

In this paper, we propose a method of judging status of light without previous light position data. Our system makes list of light bulbs registered position and status of their while running with detecting light bulbs, estimating position of their composed with 3D-model of surrounding environment. When robot patrol and check light bulbs, comparing camera image and list's data and renewing light bulb list with estimating process. Experiments performed accuracy and effectiveness of the proposed method in indoor environment.

抄録全体を表示

A compact size home robot composed of omnidirectional platform and lightweight arm is developed. Kinematic parameters of the 6-DOF arm are calibrated to increase positioning accuracy. The position of a desk in a room is recognized using depth image captured by Kinect sensor. A smooth path to the desk is computationally planned, in which clothoid curves, circles, and straight lines are connected. After reaching the desk side, objects on the desk top are recognized using both color and depth images captured by Kinect sensor. As example of object, a plastic bottle is gripped by the robotic arm. Typical total tasks of home robot to approach and handle the objects are practically realized.

抄録全体を表示

Increasing aged people who needs nursery care, shortage of care-giver is getting serious. Robot technology is one of the most promising solutions for this problem. Mobility is one of the most important functions for household works. However, it is not easy to move in houses usually, because there are inevitably obstacles on the floor (steps, clothes, etc.) and the environment is not flat. This paper proposes a simple and feasible indoor robot system that consists of a wall traversing platform and robot arm. The platform is designed to grip commercially available handle rail for adherence to the wall. The design of traversing mechanism and basic experiments are shown.

抄録全体を表示

This paper proposes a method of speaker estimation using two potential maps, human position and sound position, generated from a wide angle camera and superdirective microphones on the ceiling. When we observe natural conversation around the room for long periods, the sensor system should be constructed as a part of the room. Therefore, our method is adaptive for such situation. Using potential maps makes it able to estimate speaker position using sound data with some noise. The result of comparing speakers estimated by our method with manually generated correct speakers of an actual conversation shows an effectiveness of our method.

抄録全体を表示

This paper describes a sensor data library that collect mesurement results that captured everyday use objects. We selected dozens of objects which people need to manipulate, and collect sensor data with several different settings of sensor poses and background. 40 series of objects were selected, and color images, range images, and reflection intensity images were captured. We aim to make the sensor data library to use recognition function mounted on compact robot that assists human daily life.

抄録全体を表示

This paper proposes a robot which can pour liquid from bottle. The robot consists of a 7-axis robot arm and 4-axis robot hand, a high-speed vision, and the robot is controlled based on visual information. The robot system has the abilities to pour liquid smoothly and exactly, is independent of target. A system designer can easily adjust for target by setting several specified parameters. In this paper, first, pour motion design for pour exactly is proposed. This motion plan consists of 3 step which are extracted from human motion. Secondly, vision processing to get liquid surface information. Thirdly, drop position control using vision feedback. Finally, the data of experiments are shown and the effectiveness of the propose method is verified.

抄録全体を表示

In recent years, Japan is facing a problem of an aging society with a falling birth rate, and the number of elderly people living alone is increasing with the problem. To guard against social isolation and to connect another person, some adequate ways to ensure linkages to people who care are promising. Our solution for the problem is to propose a system based on simple interactions using a house cleaning robot. This paper presents the concept for developing the system, expected functions, prototype system, and preliminary experiments of robot positioning. In addition, the users' impressions of the system are finally evaluated.

抄録全体を表示

The purpose of this study is human-robot collision avoidance in passing by each other. We propose a recognition method for a human movement for collision avoidance, using a Kinect sensor. The method analyzes the human movement from the human's speed, position and posture. The robot recognizes the human's position from his/her waist position. The human's speed is calculated from the difference of his/her positions per unit of time. To recognize the human's walking direction, we used the difference of distances from the robot to both of his/her shoulders. For example, when the right shoulder is nearer to the robot than the left shoulder, the human is recognized to face to left. By applying this method to the robot system, the robot successfully recognized the human movement and passed by him/her.

抄録全体を表示

This report describes a robot-mediated task knowledge transfer framework for pick-and-place tasks. Such a task transfer is divided into two steps: (1) a person teaches task knowledge to a robot; and (2) the robot teaches the learned knowledge to another person. In the first step, the person performs a task and the robot observes it. The task treated here is to make a structure of boxes like an arch. The robot detects geometrical relationships between the boxes to make a model of the task; three relationships, On, Against, and Coplanar, are considered. The robot then executes the same task based on the learned knowledge, where the positions of boxes are automatically calculated according to their relationships. In the second step, the robot demonstrates the task to another person.

抄録全体を表示

This report describes a collaborative object search by a robot at a remote site and a human operator. The robot can autonomously search for a target object using an RGB-D camera. The robot performs the object search in the following three phases. First, the robot finds a desk and moves there. Then the robot extracts the candidate using a color. Finally, the robot recognizes the target object using SIFT features and estimates its pose. The operator examines the remote scene using a fish-eye camera on the remote robot, and gives advice on, for example, the object location to the robot. The robot then follows the advice to approach the target object. Experimental results show the effectiveness of the collaborative object search in comparison with a fully autonomous one.

抄録全体を表示

In this research, we develop a robot system that can find and pick up a specific object from a pile of objects with a grasp planning including sweeping motion thereby decreasing a necessary number of operations. We use local visual features for recognition and pose estimation of a target object to cope with partial occlusions. Pose estimation of unknown objects are also necessary for handling them. We obtain point cloud data by an RGB-D camera (KINECT) mounted on the robot head, and extract object surfaces by Euclidean clustering, plane detection, and normal estimation. The robot adaptively chooses grasp or sweep actions depending on the placement of the target objects and others. We implemented the proposed method on a dual-arm robot and applied it to the task of find a target object in a cluttered container. We confirmed that introducing sweep motions can effectively reduce the time to achieve the task.

抄録全体を表示

In everyday life, the collision between people sometimes occurs at the coroner. We need to solve the problem which is the collision at the corner between the service robot and the human to spread robots to the society. The purpose of this study is to prevent this problem. We, therefore, aim to make a robot's movement trajectory to prevent the collision. In the first report, we analyzed the human behavior, for example, walking velocity, walking path, walker's visibility at corner, distance to stop from the recognition. And, we considered the service robot's collision avoidance trajectory. In this paper, we propose and evaluate the service robot's collision avoidance trajectory.

抄録全体を表示

In recent years, elderly people's percentage is increasing with the declining birth rate and increasing aging population. Therefore, shortage of care support specialist is a problem. The human cooperative support robot which shares the desired task appropriately and performs various tasks with human is expected. We introduced International Classification of Functioning (ICF) for the evaluation of the target person and the definition of the cooperative support motions. By introducing this reference index, the robot can support the target person cooperatively with appropriate shared motions. As an example of human-robot cooperative motion, we considered the appropriate shared motions about meal assistance and clothes put on/off assistance.

抄録全体を表示

This paper focuses on making a table tennis robot cheaper and smaller. Unlike previous studies that used high-speed cameras to predict ball's trajectory, we propose using a depth camera (cheaper than a high-speed camera) to measure ball's velocity and position, and predicting ball's trajectory by using rotating ball's equations of motion that is based on aerodynamics. By using the equations of motion, we can also calculate ball's rotational states through ball's velocity and acceleration. We developed a tabletop type omni-directional mobile robot equipped with an arm and a depth camera, with consideration of cost and size.

抄録全体を表示

The aim of our research is development of a modular home robot. Task of the robot is managed by timetable. Timetable needs to estimate time that is longer than each task, because there is range in end time of task. So, we focus attention on environmental sounds associated with status and changes in environment. This paper describes implementation of breakfast table setting by a modular home robot with environmental sounds recognition.

抄録全体を表示

For the service robot which carries dishes at home or a restaurant, supporting different dishes of weight and maintaining the endpoint attitude to external forces acting suddenly on the robot body are required so as not to spill them or harm people. In the paper, it is shown in experiments using a six-joint robot arm that applying an adaptive control for the unknown gravity force-moment load at the endpoint and leaving the kinematic redundancy to the robot arm allow it to safely continue the task even in the situation without any force sensor.

抄録全体を表示

The ultimate goal of this study is to reveal influences of the head shaking motion on the moving stability, and relationships between gait and energy consumption for walking, trotting, and galloping. We have been developing a four-legged robot that has similar link length ratios with a typical horse. To realize stable gait motions, we have to investigate the best sets of gait parameters. In this paper, we focus on the following two parameters: the step length and the depth from the horizontal line. We conducted an experiment using the robot in trot gait, in which the body roll angle was measured to evaluate the robot's stabilization. The robot was hanged by four springs during the experiment to prevent it from falling down. As a result, we found that the depth from the horizontal line should be as small as possible to reduce the amplitude of the roll angle variation.

抄録全体を表示

This paper reports design and building of a four-legged robot that can walk in various gaits. We have developed a four-legged robot. However, the robot was designed for only one gait. Thus, we recalculated each joint torque required to achieve walk, trot and gallop gaits using Newton-Euler method. Based on the calculation, we chose appropriate motors, which required us redesigning of the body including the belt-pulley mechanism. The new body's dimensions were determined based on the length ratio between a forelimb and hind legs of a typical horse. As the result of redesigning, the dimensions of the body are 540 and 330 mm, which are slightly larger than our previous robot. We conducted some experiments to confirm whether each joint of the new robot can follow reference trajectories.

抄録全体を表示

In order to expanding movable range of life support robots, this paper proposes a robot having a new gait generation which is capable of moving on winding staircase. The gait uses a Central Support leg (CS leg) having a DOF in addition to four simple legs. Main futures of the gait are that stability margin is increased by walking with CS leg. Moreover, turning at narrow space is possible with low DOF (2×4DOF for four simple legs+2DOF for CS leg=10DOF). In addition, it is possible to ensure sufficient stability margin without moving the center of gravity, and performs static walking. We designed the robot having those legs to achieve this gait generation. In particular, the CS leg has a DOF of perpendicular extension which is realized by telescopic mechanism using the timing belt fixed to the leg and body.

抄録全体を表示

Legged robot is one of the ideal mechanisms for moving around muddy, irregular, and fragile surfaces. However, once it is turned over, it is not easy to recover, or is needed for special mechanisms, which cause the robot unnecessarily complex. To realize powerful yet simple legged machine that is able to self-recover from unexpected rolling, we developed a polyhedral shaped robot that has legs capable of performing tripod gait on every face. In this paper we show that this type of multi-faces tripod gait is realized by a simple gears and links mechanism driven by a single motor.

抄録全体を表示

The purpose of this study is to develop a frog-like robot that is able to avoid obstacles by jumping and walking on rough terrains by legs. There is no study that realizes both instantaneous jumping ability and precise control of leg joints in a single compact body. In this study, we refer to a musculoskeletal model of a frog and develop a leg mechanism that is designed to extend the leg instantly by a wire drive which is similar to a biarticular muscle of a frog. With this mechanism, two pins slide along with an L-form slot during wire tension, by which the joint swings to apply jumping force. We verified the mechanism by constructing an actual model and found that the joints could passively slide to the extending position by only applying tension on the wire for jumping. In addition, we calculated torque of the tip of a foot.

抄録全体を表示

Legged locomotion is suitable to move on uneven terrain. However the advantages of legged robots have not been achieved yet. In this study, a relationship between combinations of the joints and generated force or torque will be discussed. In this research,quadruped with four joints per leg is considered. These legs have a possibility to adapt a rough terrain more because of their redundancy. Redundant joint may change the direction of output force without losing the characteristics of maximum output force. After several considerations, three legs out of 81 combinations of joints are examined to simulate. Output force distributions during walking on at ground and slope are reported.

抄録全体を表示

In this paper, we describe the three legged climbing robot for inspection and maintenance of iron structures. The robot has three electromagnets at each leg in order to stick to a steel slope. It can climb the slope and trace a line on the slope by switching On/Off of the electromagnets properly. Legged robots usually have many actuators, and they are very difficult to control. In this paper, we developed a mechanism that can reduce the number of actuators with spring mechanism for link length adjustment. The length of a leg can be adjusted passively in order to satisfy a geometric constraint and also to put back the length of the legs after one stride. In this way, the robot has only three motors and can be controlled very easily. Effectiveness of the developed robot is shown by experiments.

抄録全体を表示

The ultimate goal of this study is to develop a mechanical foot with an arch mechanism similar to that of a human foot for bipedal robots. Many researchers consider that the arch mechanism of human foot plays an important role of shock absorber and controlling the body posture. Introducing the arch mechanism to real robots can reduce the complexity of control of biped robots. We previously reported that we developed a biped robot using the Chebyshev linkages with only one actuator and a robot foot with an arch mechanism. In this paper, we report the development of a mechanical foot with two parallel longitudinal arches. We built a mechanical foot and found that it reduces the roll angle amplitude of the body compared with other feet through experiments using one DOF legs using Chebyshev linkages.

抄録全体を表示

Since powerful actuators are required for legged robots in practical use, it is considered that air cylinders, which have high power weight ratio, are suitable. In the report, we design a biped robot using air cylinders as actuators. For controlling air cylinders, it is difficult to apply them for mechanism of robots, since they have not enough precise positioning. Each air cylinder is attached to two links and imitates a bi-articular muscle. By the imitation, position definition precision is increased, even if the air cylinder is used as actuators. Moreover, air cylinders have damper function, and this function can be utilized for absorbing impact from a floor. This report shows these effectiveness by some inspections.

抄録全体を表示

Our 4-legged running robot was designed with anteroposterior asymmetry in its body, so that it has stronger rear legs compared with the fore legs. This time, we aim to describe the robotic behavior and verify the validity of anteroposterior asymmetry in a robotic body using an inverted double pendulum model. If we employ a robotic model composed of two masses, the robot in the air can be described as a free motion of a two-mass system. If the dynamical robot motion is divided into several cases such as fore- leg landing, in the air, and rear-leg landing, and these cases are appropriately switched, the total system while running can be precisely written by a set of these motion equations. In this study, we have verified that the torque necessary for the leg swinging has anteroposterior asymmetry in the body using an inverted double pendulum model. Also, we have shown that robotic running can be described by these models.

抄録全体を表示

To reproduce the legs of a four-legged animal, we adopt as a leg module a telescopic articulated active suspension with the linear motion mechanism constituted by the "compression spring" and "DC motor". Moreover, it is necessary to efficiently reproduce the animal's leg motion in order to increase the moving speed of the robot. In this study, taking into consideration a four-legged running robot with an active suspension mechanism, the telescopic control of the leg with the landing decision reduced bouncing and stumbling of the tip of the leg during running, which improved the moving speed. As a result, turning to the right and left during running was suppressed, and movement speed improved.

抄録全体を表示

A bird like jumping robot using Slide-Worm Gear has been developed successfully. The Slide-Worm Gear consists of a slidable worm and worm wheel. This mechanism can select one of three action modes by restraining one of the motions. In case of worm mode, the worm slide motion is restrained and the worm turning motion is transfer to the worm wheel turning motion as usual. In case of rack-pinion mode, the worm turning motion is restrained, and the worm slide motion is transfer to the worm wheel turning motion. And in case of screw mode, the worm wheel turning motion is restrained, and the worm turning motion is transfer to the worm slide motion. This robot applied the Scott-Russel mechanism and a spring. The leg moves by the worm mode, spring is compressed by the screw mode and the robot jumps by the rack-pinion mode. A single mechanism works triple ways.

抄録全体を表示

This paper studies an effect of plastic deformation in the crashing process of a foot mechanism for humanoid robots. The previous mechanism that the authors developed had seriously large backlash and dry friction which caused plasticity against crash. The authors developed another mechanism which reduced backlash and have found that the shock absorption performance was rather degraded. They learned that the plastic deformation is more important than a variable-stiffness. By reducing dry friction with lubricant, deformation can be cancelled after unloading, so that the mechanical plastic deformation works for impact absorption and the following stable support.

抄録全体を表示

The Multi-purpose Module System (MMS), which is one of the module robots, consists of joint modules and function modules. It is possible to transform the MMS such as series-connected form and quadruped form and add appropriate sensors. Therefore, the MMS is useful as rescue robot at unknown situation. This paper presents a quadruped form MMS (MMS-G02). Especially, modeling and simulation of the MMS-G02 for stability control is discussed. An inverted pendulum model is proposed and the MMS-G02 is controlled using a state feedback control to stay in balance. And the effect of friction force between the MMS-G02 and the ground is also evaluated by simulation. The maximum value of the lean of the MMS-G02 decreases with increasing the friction coefficient. The maximum value of the lean is 0.21 rad at condition of the friction coefficient 0.1. The results also indicated that the MMS-G02 can stay in balance and keep on the two legs.

抄録全体を表示

This paper proposes a walking control method for small-sized biped robot by using trajectory planning. First, for analyzing a walking motion of biped robot, we regard it as the three-dimensional inverted pendulum. Then, we plan the trajectory of the center of mass and many foot places during walking in consideration of kinematics and kinetics in relation to the analyzing result. By using the trajectory result, a walking control system for small-sized biped robot is configured. And to confirm the effectiveness of the control system, experimental results of walking control of the 12 d.o.f biped robot are shown.

抄録全体を表示

In this paper, we develop a small amphibious legged robot which travels on rough terrain, passes through narrow space, and runs on water surface. Although many robots have been developed to observe and investigate in a natural environment, rough terrain, bushes and swamp have disturbed their travel and operation. Here, we focus on a scale effect advantage that the smaller, the larger the area force against the volume force and manufactured a reduced degree of freedom hexapod robot with the body length of 5.2cm and the weight of 7.9g. As a result of on-water experiment, the robot with the specific gravity larger than 1 achieved the payload of 11g (116%) on water by the paddling of 19Hz.

抄録全体を表示

Many moving robot to search in uneven ground has been developed. However, carrying robots in uneven ground is dangerous. Operators may injure him during carrying in. If robots are able to throw in uneven ground, it is safety and easy to carry robots. The purpose of this study is to develop a durable robot. Thus, we developed quadruped robot with spherical shell. And for the realization of rotational motion using a spherical shell is the purpose of this study was to calculate the joint angle of the appropriate leg from the coordinates of the feet in the posture of the target by solving the inverse kinematics.

抄録全体を表示

Many moving robots for search in uneven ground has been developed. However, carrying robots in uneven ground is dangerous. Operators may injure him during carrying robots in uneven ground is dangerous. Operators may injure him during carrying in. If robots are able to throw in uneven ground, it is safety and easy to carry robots. The purpose of this study is to develop a durable robot. Thus, we developed quadruped robot with spherical shell. Further, solve problem of the prototype and add the rolling motion by infinite rotation of the servo motor. Sliprng makes it possible to perform an infinite rotating without entwined cables.

抄録全体を表示