We developed a fingertip-size proximity sensor that detects the distance to and the tilt angle of the surface of an object with the peak-to-peak distance error less than 1/129th (<;31 μm) and the measuring time less than 1/10th (<;1 ms) those of existing sensors. In addition, we realized the task of catching a fragile object (paper balloon) with a high-speed robot hand equipped with the sensor. High-speed, high-precision sensing enables contact detection independently of the contact force. We describe a robust contact detection method that does not depend on the reflectance, the tilt angle, or the shape of target objects surface. In experiments, we confirmed that the sensor could measure a distance of 2.85-20 mm (resolution: 44 μm in <;3 mm) and a tilt angle of ±45° (error: 1.47°) for objects with reflectances of 18% and 90%. We demonstrated that the hand could catch the paper balloon when dropped from a height of about 20 cm without crushing it.
25 January 2021 in IEEE Transactions on Instrumentation and Measurement
This study proposes a visual tracking system with 500-fps image processing that can detect and track a fast-moving appearance-varying target in a 360-degree-field of view (FOV) from high-spatial-resolution images. The system consists of a high-speed camera, a pan-tilt Galvano-mirror system, and a mirror-drive high-speed pan control system with 360-degree-rotational capacity. Omnidirectional visual tracking of a remote object using complex visual features becomes possible due to the acquisition of high-spatial-resolution images of the target object via the system. We introduce an intelligent high frame rate (HFR) target-tracking algorithm that can simultaneously detect and track an appearance-varying object against a complex background for high-speed visual feedback control. Furthermore, the proposed surveillance system enables robust visual tracking under low-light conditions using a 3-D light detection and ranging (LiDAR) sensor system and a simultaneously controlled active spotlight projection system. The effectiveness of the proposed omnidirectional visual tracking system was demonstrated by conducting several omnidirectional detecting and tracking experiments on persons dozens of meters away using a 25-mm focal length lens under both daytime and night-time conditions.
7 January 2019 in IEEE Robotics and Automation Letters
In this paper, we propose a combination of sensing and control modules for catching soft objects (i.e., a marshmallow and paper balloon) at a high speed with small deformation. A paper balloon and marshmallow are deformed by a small force and they have individual differences in terms of softness. Therefore, model-based, small-deformation catching is difficult. To realize small-deformation catching, we used high-speed sensor-based (vision and proximity) control modules without a deformation model of the soft objects. A high-speed vision-based controller adjusts the positions of the fingertips of a robot to some extent. As the distance to the object decreased, the fingertip positions were accurately controlled by the high-speed, high-precision proximity-based control. Furthermore, the fingertips were stopped by a proximity-based contact detection before the object surface was deformed. Virtual damping control was effective in catching an object whose surface can be easily deformed by a small impact force or vibration of the fingertips, such as a paper balloon.
Vision-based structural displacement methods allow convenient monitoring of civil structures such as bridges, though they are often limited due to the small number of measurement points, constrained spatial resolution, and inability to identify the acting forces of the measured displacement. To increase the number of measurement points in vision-based bridge displacement measurement, this study introduces a front-view tandem marker motion capture system with side-view traffic counting to identify the force-inducing passing vehicles on the bridge’s deck. The proposed system was able to measure structural displacement at submillimeter resolution on eight measurement points at once at a distance of 40.8–64.2 m from a front-view camera. The traffic counting system with a side-view camera recorded the passing vehicles from two opposing lanes. We conducted a 35-min experiment for a 25 m-span steel road bridge with hundreds of cars passing over it and confirmed dynamic displacement distributions with amplitudes of several millimeters when large vehicles passed.
1 January 2021 in IEEE Robotics and Automation Letters
In this letter, we propose a fast mirror-drive pan-tilt target tracking system that can robustly track an object whose appearance varies in a complex background at 500 fps. By assuming a small image displacement between frames, which is a property of high-frame rate vision, we develop an fast object tracking algorithm by hybridizing the convolutional-neural-network (CNN) based object detection with template-matching (TM) based tracking operating at hundreds of frames per second (fps). For object tracking with high-speed visual feedback, the proposed tracking algorithm can remarkably reduce dozens-of-milliseconds-latency in the CNN-based object detection by simultaneously executing TM-based tracking for several images at consecutive frames within a few milliseconds. In the proposed pan-tilt tracking system, when the current tracked objects are occluded or out of the camera view, it can recognize objects to be newly tracked with CNN-based object detection at the rate of 33 fps with acceleration using graphic processing units (GPUs). Controlling the pan-tilt tracking system via visual feedback at 500 Hz, fast moving objects can be robustly tracked at the center of the camera view. The effectiveness of our method was experimentally demonstrated via several results when fast-moving pre-learned objects, such as toy cars were tracked in complex backgrounds.
7 July 2016 in IEEE Robotics and Automation Letters
In this study, the impedance control for smoothly receiving the impact of an incoming object is designed and realized. The control strategy is based on the idea that the back drive motion of an end effector attributable to the collision impact is regarded as plastic deformation of the robot. The impedance dynamics are constructed from the Maxwell model, which describes plastic deformation. Next, two types of control methods are proposed in terms of the connection configuration of the spring and damper. Physical simulations of the impact absorption with a robotic arm are executed to validate and analyze the proposed control laws.