This paper presents a dynamical system based approach to assistive teleoperation for object grasping. According to the target position and the current user input, a virtual desired goal is generated for a stable second order linear dynamical system via linear interpolation. In order to achieve collision avoidance between the target object and the robot finger, a virtual force is generated as an input to the dynamical system for online trajectory modification. Experimental results of an object grasping task are presented to demonstrate the effectiveness of the proposed assistive approach in a simulated robot teleoperation system.

This paper proposes an imitation learning-based approach to assistive teleoperation for a reaching task with multiple objects. First, according to the approaching velocity of the user's input, we estimate the user's intention of the target object. Then, by integrating user input and the output of the demonstrated trajectroy via Dynamic Movement Primitives, we generate the robot endeffector trajectory that will converge to the position of the target object. We present experimental results of an object reaching task to demonstrate the effectiveness of the proposed method in a simulated robot teleoperation system.

This paper describes the drone-based visual assist system for the teleoperated construction robot. The drone hovers at the side or front of the construction robot to capture images of the work area. The operator switches the drone's hover position by operating a switch. If many hover positions are set, the operator can determine the drone position in detail, but the operational burden may increase. This paper proposed several combinations of hover positions and compared them experimentally.

In a robotic system to pack food in a bento box，robots need to handle a wide variety of food with different physical properties and indefinite shapes. In particular, grasping and high-speed transporting of such food is required. In this paper, we propose a search method of stable grasping points for this purpose, and verify whether it is achieved with the grasping points derived by the proposed method through experiments on different kinds of actual food.

To observe the inside of gas pipes, we have been developing a pneumatically driven inspection robot that can locomote in service line including bended areas. In this paper, the robot accomplished insertion from a meter gas cock and shortage of the transmit time. To enable it to insert from a meter gas cock, the diameter of the robot is decreased to 12 [mm]. Furthermore, to shorten the transit time, we developed a way to increase the airflow rate of the robot by applying a wider tube stored in a helical shape inside the body. Finally, the robot could insert from the meter gas cock throughout a 7[m] service line within 81[min].

In the face of an aging society and a shortage of young labor, the lack of young workers on construction sites has become increasingly challenging. Various types of construction robots have been developed to handle construction tasks. Among them, material transport robots can transfer heavy materials or tools to the desired location. However, most of the existing studies have focused on planner motion and have not discussed the performance and possible challenges of mobile robots on steps or slopes on construction sites. In this work, we conduct force analysis of the robot on a slope and develop a motion planning method accordingly so that robots could act as human operators. With the proposed method, the mobile robot can move more efficiently on the slope without upgrading the hardware or changing the environment. As a result, the mobile robot can perform tasks on construction sites more freely and efficiently. The effectiveness of the method has been evaluated on a construction site with the task of entering an elevator by climbing a slope.