We proposed a picking robot system which is apllicable to various mixed items in shelves. The robot has a two-finger gripper which can change the open width of the finger. To determine the position, the pose and the open width when the gripper pick items, we proposed efficient determination algorithm which is based on a RGBD sensor data. In our experiments, 25 items of Amazon Picking Challenge 2015 can be picked well by our proposed system. In this paper, we describe the system, the algorithms and the experimental results.
This paper presents the application of our passivity-based whole-body contact force control framework to compliant walking. We experimentally evaluate the performance of the controller using our original torque-controllable hydraulic humanoid robot, TaeMu. The experiments are performed on the postural stabilization against support space tilt disturbance and walking on the uneven and unstable ground. Instead of limiting the walking speed slow, we address the terrain adaptability of the proposed robot system and algorithm. For that purpose, prior terrain information and environment perception are not given to the controller. The desired motion trajectories are given in advance on the assumption of the ground to be flat. Nevertheless, the robot was able to compliantly keep its posture, compliantly walk on small steps and on an unstable rocker board laid lengthways or crossways. The results demonstrate the effectiveness of the proposed controller as well as the robotic hardware.
In Japan, deterioration of many tunnels and bridges have become a serious problem. Moreover, engineers that manage them are insufficient due to aging. Therefore, we developed the hammering robot that can imitate hammering sounds of inspection workers. When we use this robot, workers can detect concrete defects by using their experiences. For example, if we attach a video camera or microphones to this robot, they can detect defects as before at remote locations. Furthermore, it can contribute building high accurate automatic inspection systems by learning hammering sounds of inspection workers. In this paper, we described these systems using this robot. To verify usefulness of these systems, we conducted experiments using a concrete test block and compared the results of an inspection worker with this robot. As a result, we confirmed that this experiments showed the results of this robot is more useful than its of an worker.