In this paper grasping of a dynamically moving object is achieved using high-speed visual and force feedback. First a hierarchical processing model based on high-speed sensory feedback to realize manipulation in a changing environment is proposed. In this model high-speed sensory feedback is used at hierarchical three stages: (1) servo control, (2) motion planning, and (3) switching of motion planners. Because the motion planners are switched according to an external environment at a high rate, responsiveness and flexibility to dynamic changes of the environment is achieved. Next based on the proposed model a manipulation algorithm using high-speed visual and force feedback is proposed, in which a grasping task, an object handling task, and a collision avoidance task are executed. At last experimental results are shown based on a sensory-motor fusion system with a dextrous hand-arm and a high-speed active vision.
A new approach to the forming control using forming process model of rheological objects is presented. Manipulative operations of rheological objects can be found in many industrial fields such as food industry and medical product industry. Automatic operations of rheological objects are eagerly required in these fields. Since rheological objects deform during operation processes, it is necessary to estimate their deformation for the automatic operations. We will propose a forming control of rheological objects using forming process model. First, an extensional forming system of rheological objects is developed. Second, we will propose a forming process model consisting of the expanded shape and the residual part. Next, we will propose forming control using forming process model and will demonstrate the validity of the proposed forming control experimentally.
In this paper, a haptic device for multi-fingers is proposed. The feature of this device is as follows. (1) This device consists of a probe, several sets of force controlled manipulator, and a trajectory controlled manipulator. (2) The bases of the force controlled manipulators with small links of low inertia are attached to a tip of the trajectory controlled manipulator. The tips of force controlled manipulators are attached to an operator's fingertip, respectively. (3) The base of the probe is also attached to the tip of the trajectory controlled manipulator. The tip of the probe is attached to the operator's hand. Thereby, the probe continuously and correctly detects the position and the orientation of the operator's hand. (4) The trajectory controlled manipulator moves the tip so that the force controlled manipulator's base may become suitable position and posture to the operator's hand. (5) Since the force controlled manipulator's bases are grounded through the trajectory controlled manipulator, the force controlled manipulators can generate suitable haptic sensation to the operator's fingertips.
This paper proposes a novel manipulation method with a quadruped robot using its whole body. The quadruped robot we have developed can take various postures. Two of its legs can support its body by standing on the knees and the other two can serve as arms. The robot can hold a relatively big and heavy object with its two arms and can manipulate the object by moving its body. The two legs standing on the knees form a closed link system. Its desired configurations during the manipulation are at an increased mobility singularity if all knees and tips of the two legs are in contact with the floor. The problem with the singularity is that the mobility of the robot changes significantly around it. This paper shows a contact condition for the legs free from such a singularity. We conduct experiments to verify that our quadruped robot can perform the manipulation. A possible application of the manipulation is loading/unloading an object onto/from the body of a quadruped robot.
In the case that the robot arm has redundancy, its ellipsoid of manipulatability doesn't give enough information. In this paper, we study the relation between ellipsoid of manipulatability and mapping of cube which means range of possible angular velosity by Jacobian matrices. The mapping is a polyhedron. It, means all possible velocity of the robot arm's end effector. The polyhedron has no lack of information of its manipulatability.
Visual attention is one of the most important issues for a mobile robot to accomplish a given task in complicated environments since the vision sensors bring a huge amount of data. This paper proposes a method of sensor space segmentation for visual attention control that enables efficient observation taking the time needed for observation into account. The efficiency is considered from a viewpoint of not geometrical reconstruction but unique action selection based on information criterion regardless of localization uncertainty. The method is applied to a four legged robot that tries to shoot a ball into the goal. To build a decision tree, a training set is given by the designer, and a kind of off-line learning is performed on the given data set. Discussion on the visual attention control in the method is given and the future issues are shown.
An optimal control problem for nonlinear systems is useful because various control objectives can be included in the cost function such as rapid stabilization and input cost minimization. However, there exist input restrictions of actuators and state constraints that arise from the movable region of the system in the real system. Therefore, the optimal control problem should be solved to satisfy these restrictions. In this paper, we propose a new method to realize nonlinear receding-horizon control under input restrictions and state constraints. First, a receding horizon control problem is formulated. Next, a new real-time optimization method is proposed. Then, the method is applied to solve a receding-horizon control problem of a wheeled vehicle. The effectiveness of the proposed method is shown through simulations and experimental results.
Electromyogram (EMG) has been often used as a control signal of a prosthetic arm, which includes information on not only muscle force but operator's motor intention and mechanical impedance of joints. Most of previous researches, however, adopted the control methods of the prosthetic arms based on the EMG pattern discrimination and/or the force estimation from the EMG signals, and did not utilize any knowledge on tasks performed by amputees such as a grasping-an-object task and a spooning-soup task. In this paper, a new EMG pattern discrimination method is proposed using a statistically organized neural network and an event-driven task model. The neural network outputs aposterioriprobabilities of motions depending on the EMG signals. The task model is represented using a Petri net to describe the task dependent knowledge, which is used to modify the neural network's output. Experimental results show that the use of the task model significantly improves the accuracy of the EMG pattern discrimination.