In this paper, internal constraints based control is proposed as an autonomous control method to realize both cooperativity and diversity in multiple robots. We used a robot model composed of two subsystems. One is a physical system which selforganizes a dynamical coherence in mutual interaction between the robots and their environment. The other is an interpretational system, and it generates internal constraints from the coherent state and operates the physical system based on the constraints. The proposed method is realized through mutual constraint process between these two subsystems. To check the effectiveness of this method, a task that multiple robots carry one load was examined as an example of cooperative behavior.
The concept of discrete-time nonholonomic systems, in which the constraints cannot be represented as algebraic equations of generalized coordinates, is introduced. In particular, the constraints formulated as differece equations of generalized coordinates are considered. Such systems can be seen in the digital control of continuous-time nonholonomic systems, and in mechanical systems with repetitive and discontinuous constraints. A two-wheeled mobile robot and pivoting manipulation of a polyhedral object are described as simple examples. The k-step reachable region is defined as the set of the k-th state which the system can reach from the initial state, and the reachability of such systems is discussed. A motion planning method using the Jacobian matrix of the state with regard to the input series is proposed.
In this paper, an in-pipe mobile robot is developed, which is composed of two types of soft actuators. One is a traveling actuator that can be expanded to its axial direction, the other is a holding actuator that can be expanded to its radial direction. These actuators are made of a silicon rubber and are driven with a pneumatic power. The holding actuator can adaptively change its shape owing to the flexibility of the silicon rubber and the air pressure. Therefore, the robot can travel in various types of pipes, such as changing diameter at a joint, wiring in a pipe, and pipe with an arbitrary sectional shape. Because the whole body of robot is elastic, this robot doesn't make a damage for the pipe wall and can travel in an elastic pipe. Because the robot is simply structured, it is easy to be scaled down. These features are expected for traveling in a living body. The influence of the tube length (from servo valve to robot) is compensated by an improved control system. The compensator composes the transfer function that is identified the pressure response of robot and tube. In case of the pipe has fluid flow, the holding actuators are added the passage pipe, this robot can travel in such a pipe. The validity of this robot is confirmed through some experiments.
We have found a difficulty in the learning of a human-friendly behavior in a face robot by means of human instruction. The difficulty is mainly spotted at the learning algorithm of the robot not taking account of characteristics of human instruction. This paper suggests an effective learning algorithm constructed by the results obtained through analysing characteristics of human instruction, and evaluates its effectiveness by computer simulation and human-face robot interaction experiments.
We will propose a new control method called a variable constraint control which uses the first integral as an invariant manifold obtained by adding controlled constraints to nonholonomic mechanisms. We apply the results to the posture control problem of free flying robots and investigate relations between the convergence of the control and structual parameters as well as initial conditions of robots.
In this research, we discuss motion control of link system which is mounted on a foundation with an inclination of unknown angle. Here, the most interesting point is that the link system is assumed to have only internal sensors instead of external sensors. That is, the system can not measure the absolute posture (the direction of the vertical direction) . In this situation, we try to control the absolute posture of the link system. In this paper, constructing a nonlinear observer for estimating the absolute posture of the link system, we design an exponential stabilizing control law. The notable point of this paper is the simplicity of the result. This simplicity is due to the special features of the system.
This paper describes the autonomous navigation system of an unmanned helicopter, in which GPS information and image information are integrated and the unmanned helicopter R-max produced by Yamaha Motor Inc. is used. The developed system can track a moving target based on pyramids images generated by an image splitter. It can explore a wide area and re-capture a target when it is lost by means of active vision concept. During navigation, the navigation control system manages flying modes according to an assigned task and controls the speed of the helicopter. To show the performance of the developed system, experiments of searching and tracking a truck moving on ground are performed.
Minimally invasive cardiac surgery is getting strong recent attentions in medical robotics. In particular, Minimally Invasive Direct Coronary Artery Bypass (MIDCAB) requires surgeons the precision of hand skill and the mental concentration, since it needs to work on beating hearts. In this paper, we propose a surgical robot system that compensates motions of organs during operations. The motion canceling robot system includes two technologies; visual stabilization and motion stabilization. The visual stabilization provides the surgeon with the image of stabilized target point on the monitor. The surgeon operates the master robot referring to the stabilized image. The motion stabilization, on the other hand, transforms the master motion and controls the slave robot being synchronized with the heart beat. The concept of system is proposed and its prototype is designed and developed based on the frequency analysis of the heart beat. The in vivo experiments verify the effectiveness of the technologies.
Tracking of a feature point in profile data of the workpiece's joint, using a laser range finder, is indispensable for automating path-tracking tasks, i.e. welding, sealing, deburring, and so on. In this paper, a tracking method is developed that achieves higher work quality and enables looser constraints on the shape of the workpiece. In this method, a profile is recognized by minimizing energy defined by the relationship between measured profile-data points and an articulated object model (AOM), that adjusts to the various changes in profile data by modeling a workpiece's corners, surfaces, and interstices between workpieces. Experiments using a workpiece's profile data measured with a scanning range finder are shown that the proposed method tracks the feature point robustly, even in case of the complicated changes, such as shape transformation from concave to convex, interstice variation, and shape break-down due to obstacles. Moreover, experiments using simulated profile data confirm that the proposed method is robust towards the environmental changes.
The purpose of this research is to make a machine or a robot jump highly using small output actuators. A jumping machine with a self-energizing spring system is introduced, which is designed to jump using small actuators and springs. Power for the self-energizing system is generated using small actuators. And it can jump. The machine can jump or increase the jumping height in jumping phase when the motion is reasonable. The motion control trajectory of the angle of body link is introduced in this paper. Computer simulations and experiment are performed to prove the validity of the concept of jumping method by means of the self-energizing system.
This paper describes PC-based, robust, real-time disparity image and flow image generation system, which is able to be embedded in robot's body. Real-time feature is carried out in three steps: adopting recursive correlation method, optimizing algorithm to second CPU cache, utilizing MMX instruction set. Consistency checking method is adopted to achieve robustness, we integrate into recursive correlation loop to achieve realtimeness. Finally evaluation of proposed method is denoted.