This paper deals with the problem of controlling a spherical rolling robot that has a new driving mechanism equipped with a gyro. The translational motion of the robot is caused by transmission of angular momentum from the gyro to a spherical outer shell of the robot. A feedback controller that makes the robot achieve a straight motion is designed based on the equations of motion for the robot. The effectiveness of the controller is demonstrated by experimental results.
This paper proposes the human body motion interface for personal mobility vehicle including the user's twisting motion. First, saddle is attached on the personal mobility vehicle using the seat post which has the universal joints at the attachment portion. The universal joints have three rotational joints where the potentio meters are attached on each. By hip motion, the joints are moved. The potentio meters detect these rotations. Next, we introduced the sigmoid function to connect the motion of hip and the velocity of the personal mobility vehicle. Finally, the experiment was conducted to confirm the usability. In the experiment, the control interface was prepared which doesn't use the twisting motion of the hip. The subjects drive the personal mobility vehicle on the figure 8 course layout by using proposed interface and control interface. The lap times were measured in both interface and compared. After driving the vehicle, the paired preference test was conducted. The lap time using proposed interface was shorter than the control interface, and the paired preference test showed the proposed interface is intuitive.
In this paper, a human detection method that combines information of segmented range image and human detectors based on image local feature is proposed. The method uses a stereo vision system called Subtraction Stereo, which extracts foreground information and range image. Range image is segmented for every object by Mean Shift Clustering. The proposed method reduces processing time and false detection by limiting the search range to the object. Joint HOG feature is used for human detection and reduces undetected human by integration of detection windows in consideration of occulusion. The proposed method is evaluated by experiments comparing with the method using the conventional Joint HOG feature.
This paper reports experimental results on a surgical grasping forceps that equips a vibration actuator to enhance a tactile perception ability. A short-time exposure of tactile receptors to sub-sensory white-noise vibration is known to improve perception ability. This phenomenon, called stochastic resonance (SR) in the somatosensory system, is expected to enhance the sense of touch when the weak vibration is applied, and thereby improve associated motor skills. To investigate the effect of the noise intensity, a summing network of FitzHugh-Nagumo model neurons was built. The simulation results showed that appropriate noise intensity improves the coherence between the input stimuli and the network response. To confirm the efficacy of the proposed method on the surgical grasping forceps, passive and active sensory tests have been conducted. A lead zirconate titanate (PZT) actuator was attached on the grip of the forceps to generate the physical noise. The experimental results show that the appropriate noise improves the detection capability of the stimuli.