This paper describes a local path planning method of a mobile robot to search path in an unknown environment by using visual information. The mobile robot system has a hierarchical path planning system which searches path efficiently in an uncertain environment. The planning system consists of a global planner and a local planner. The global planner gives a global path in termes of a sequence of visual sub-goals. Then the local planner generates a local path between the sub-goals with the help of visual sensor. The main focus of this paper is on the local path planning which provides real-time guidance to the system. Visual sensor can provide useful information about the environment. So an algorithm is proposed to generate avoiding points by using visual information to bypass unknown obstacles in the local path planning. The local path planning in simple environment is simulated by using three dimensional graphics. A simple experiment is also done for the case that there are two obstacles. The validity of the proposed method is verified by these simulations and experimental results.

View full abstract

This paper describes a navigation method of a mobile robot based on cooperation of sensors. Among various sensors, vision is expected to play a very important role in control of intelligent mobile robot. In this paper firstly a method is presented to guide the robot in real time by using visual sensor. The mobile robot is controlled to follow visual sub-goal by visual feedback. The robot can detect unknown obstacles and make avoiding paths automatically. Obstacle avoidance based on visual information is then simulated. The result of the simulation shows that navigation using only visual information has several problems, because it is impossible to get the environmental information close to the robot by visual sensor. In order to solve those problems, visual sensor and range sensors such as ultra-sonic sensors, are cooperated together. It is important to integrate the information of multiple kinds of sensors. A control method, which utilizes the information fusion between vision and range sensors, is proposed. The validity of the method is verified by simulations and experimental results.

View full abstract

Since the characteristics of flexible robot arm motion is far more complex than that of rigid arm motion due to its link elastic deflections, the flexible arm end-effector positioning problem also becomes more complex. The problem is finally resolved into the following three sub-problems: (1) how to suppress the link elastic vibration, (2) how to achieve accurate joint positioning, and (3) how to compensate the end-effector positioning errors due to the link deflections. The problem (1) is being solved by many pieces of work. The problem. (2) arises also in the case of rigid arms but, since the joint positioning and link vibration suppressing are coupled, it becomes more complex for the case of flexible arms. The problem (3) is important in order for the arms to perform tasks but no effective method has been presented so far to solve it. This paper presents a hierarchical control system which incorporates organically three control functions: joint positioning, link vibration suppression, and end-eff ector positioning error compensation. The convergence condition for the compensating control is derived theoretically for the condition of static gravitational loads. The effectiveness of the proposed control system is proved by experiments using a two-link flexible arm. The link deflections are measured by a newly devised and developed sensor consisting of a semiconductor laser and a position sensitive detector (PSD).

View full abstract

Recently, several articulated manipulators are becoming applied to the ultimated world of nuclear plant, ocean, disaster place and space. Therefore, barous advanced manipulators are required in these environments.
Especially, in case of the manipulators under the water, the difficult problem occures, that the water disturbance prevents the manipulator from moving precisely.
In this paper, the hybrid position/force control with sway-disturbance compensator of the underwater manipulator is presented.
In the first place, the hybrid postion/force control system of the 6 d. o. f. manipulator is proposed. Because this control method is executed without the calculation of Jacobian inverse matrix, this controller can naturally return to the well-known position servo controller.
In the next place, the control compensator with feedforward signals of sway-disturbance and feedback signals of end-effector sensor is proposed.
In the final place, the integrated control system of both the hybrid position/force control and the sway-disturbance compensator is constructed with 32-bit micro-processors, and several experimental results of the 6 d. o. f. underwater manipulator are presented.

View full abstract

This paper proposes a laser acquisition control method for a corner cube reflector (retroreflector).
We have been developing an active positioning system for roboted vehicle. This system is composed of vehicle-borne laser scanner and corner cube set on environment. The on-board laser scanner controlls the emissive direction of a laser beam so as to usually point to the corner cube. During this control, namely tracking control, the vehicle measures the directions of the beam and determines the own position by the trangulation principle.
In order to realize the positioning system, we have proposed a tracking control method by actively using the reflectional characteristics of corner cube, and performed the high accurated tracking. However the method has a disadvantage that the laser beam cannot reentry tracking if the beam once misses the corner cube.
This paper presents an acquisition control method which searches and captures the corner cube in order to continue the tracking. The acquisition control method is simply performed by using a fan-shaped laser beam. The experimental results using proto-type system show the effectiveness of the method.

View full abstract

It is quite desirable for an operator at the remote controls to have a real-time sensation of presence as if he or she were in the remote anthropomorphic robot and to be able to maneuver it dexterously.
This concept is called TELE-EXISTENCE. Realization of a visulal display with a sensation of presence is one of the most important elements of this tele-existence. In this paper a method is proposed to realize a tele-existence display and its design procedure is explicitly defined. Experimental display hardware was made, and the feasibility of the visual display with a sensation of presence was demonstrated by psychophysical experiments using the test hardware.

View full abstract

In recent years, space has attracted special interest as a new application field of robotics. A robotic teleoperator system installed with space manipulators will play an important role in future space projects, such as constructing space structures or servicing satellites. However, in space environment, the lack of a fixed base arises many problems in controlling space robotic systems. In general, any motion of the manipulator arm will induce reaction forces and moments which disturb position and attitude of the supporting base satellite. To establish a control method for space manipulators taking dynamical interaction between the manipulator arm and the base satellite into account, the authors investigate the kinematics of free-flying multi-jointed link systems by introducing the momentum conservation law into the formulation and derive a new Jacobian matrix in generalized form for space robotic arms. By means of the new matrix, they develop a control method for space manipulators based on the resolved motion rate control concept. The proposed method is widely applicable for not only free-flying manipulation problems but also attitude control problems. The validity of the method is demonstrated by computer simulations with a realistic model of robot satellite.

View full abstract

Postural control of a monoaxial bicycle which has a twin wheel and an upper structure supported on its driving axis has been studied. The bicycle was modeled as an inverted pendulum pivoted on the wheel axis. A controlling moment, that is, a restoring torque on the inverted pendulum is applied by controlling the rotation of the wheels. A control system of the model was designed according to the proportional feedback control scheme in considering time delay during the sampling and the control. Feedback gains was derived from Routh-Hurwitz's stability criteria. Experiments were carried out using these feedback gains, and the postural control of the developed monoaxial bicycle has been attained. Furthermore, driving control of the vehicle through both method--gain changing and servo-reference methods--has been achieved by applying the results obtained in the postural control.

View full abstract

In this report, the numerical simulation of the horizontal articulated robot consisting of two horizontal rotating links and a vertical linear link is presented. In the numeriacal simulation, the pendulum vibrations of the vertical linear link, in the plane perpendicular to the second horizontal rotating link, are considered. Then, in the derivation of the state equation of the robot, the equation of motion of the mechanical system is derived in considerration of the flexibility and viscosity of the coupling mechanism between the second horizontal rotating link and the vertical linear link as well as the nonlinear flexibility of the harmonic drives, and the nonlinear state equation of the robot is obtained by coupling the equation of motion with the electric current equation of the servo position control system. based on the proportional plus integral plus derivative control. Furthermore, the numerical simulation results are demonstrated, and the effects of the pendulum vibrations of the vertical linear link on the dynamic characteristics of the robot and the control accuracy are investigated.

View full abstract

This paper presents a general and efficient method using a configuration space for planning a collision-free path among known stationary obstacles for a manipulator. The basic approach taken in this method is to restrict the free space concerning path planning and to avoid executing unnecessary collision detections, based on the idea that a collision-free path can be planned using only partial information of the configuration space. The configuration space is equally quantized into cells in a six dimensional array, and the cells cencerning path planning are enumerated by simultaneously executing multiple search strategies. The coefficients for heuristic functions are randomly generated, and search strategies of different characteristics are defined for enumerating free space cells. The efficiency of each search strategy is evaluated during free space enumeration, and a more promising one is automatically selected and is preferentially executed. The total number of necessary collision detections for free space enumeration mainly depends on the most efficient search strategy among these ones. Therefore, the free space cells are efficiently enumerated for all kinds of manipulators of different kinematic characteristics in all kinds of working environments. This method has been actually implemented and has been applied to several examples which have different characteristics.

View full abstract