Robots that are supposed to work and communicate with humans in our daily life have been developed in recent years. For this kind of robots, sensitive tactile sensors are necessary to achieve haptic interactions. However, noises on tactile sensors caused by a robot’s own motions sometimes become larger than the signal to be detected. This paper proposes a method to reduce such noises. This method is realized by estimating and subtracting noises from sensor outputs. The noises on tactile sensors are estimated by using the sequence of joint angles of the robot. Our method builds upon a partially linear model to estimate noises on a tactile sensor. The robot’s posture space represented by its joint angles is divided into several subspaces to fit to a linear model. We conducted an experiment with a robot covered with tactile sensors to verify the validity of our method. This paper shows that the robot is able to estimate noises on a tactile sensor based on the proposed model.
To overcome the difficulties associated with minimally invasive surgery, we have developed a magnetic resonance (MR)-compatible compact surgical robotic system. This system uses MR-guided navigation and can augment the surgeon’s eye-hand skills that are limited by endoscopic surgery. To achieve MR-compatibility of the manipulators, we had to provide remote actuation of joints from the ultrasonic motors located and driven at least 750[mm] away from the center of the MRI. We then developed a new surgical instrument and a novel tube-rod transmission mechanism comprised of a flexible PEEK® rod encased in a Teflon® tube, as well as a technique for controlling the transmission mechanism. It was found that stiction force had a significant impact on the stiffness of transmission using a parameter estimation method. We derived an algorithm to compensate for the expansion of the rods that occurred so we could negate the stiction force. The result of the compensation showed that the time delay between the reference to the ultrasonic motor and the movement of the joint was reduced from 600 to 80[ms]. Using the master-slave manipulator system with the tool and the transmission mechanism, we were able to perform each primitive motion of a suturing task for training material.
The passive walker with knees can naturally execute the leg motion, which is essential to take a step forward, by the dynamics of legs under the gravity only. On the other hand, an inadequate dynamics causes an undesirable leg motion, such that the stance leg bends at the knee joint, the swing leg unsuitably strikes its foot on the slope and the walker falls backward by the negative gravitational moment. Though an arc foot is very important for the leg motion, its dynamical effects and mechanism have never been clarified, and a useful design framework of its shape has never been established. In this study, for the sake of simplicity and clarity, the linearized simplest walking model is used. In this paper, the dynamical effects of the arc foot, which can keep the knee joint of stance leg straight only by the stopper and can enhance the flexion of knee joint of swing leg, are discussed. Also, its geometrical negative effect is examined. Furthermore, the forward-falling condition of the walker is derived based on the energy analysis. Finally, the desired arc foot is designed and the actual effects are confirmed by a walking experiment.
As robots become more ubiquitous in our daily lives, humans and robots are working in ever-closer physical proximity to each other. These close physical distances change the nature of human robot interaction considerably. First, it becomes more important to consider safety, in case robots accidentally hit the humans. Second, touch feedback from humans can be a useful additional channel for communication, and is a particularly natural one for humans to utilize. Covering the whole robot body with malleable tactile sensors can help to address the safety issues while providing a new communication interface. In this paper, we discuss attempts to solve some of the difficult new technical and information processing challenges presented by flexible touch sensitive skin. Our approach is based on locality of haptic features for classification of touch interactions. We hypothesize that useful haptic features are composed from local sensor output pairs. We found that using sparse sensor pairs containing as little as 15% of the full sensor combination set it is possible to classify interaction scenarios with accuracy up to 80% in a 15-way forced choice task. Visualizations of the learned subspaces show that, for many categories of touch, the learned sensor pairs are composed mainly of physically local sensor groups as we hypothesized.
A conventional floor polishing mobile vehicle or robot that has a rotary brush should be fairly large in body size and heavy in weight to prevent unnecessary movement caused by friction between a rotating brush and a floor. When those robots having larger weight are used to sweep or polish a floor, accidental collisions may damage office fixtures in a room. From this point of view, a light weight polishing robot is preferable. To satisfy this demand, the floor polishing robot using a rotary brush polisher was developed. But it has danger to roll up an obstacle. This paper shows a new control scheme for the floor polishing robot, which has two rotary brush polishers. This omnidirectional mobile robot requires no driving for locomotion and steering because frictions between a floor and rotating brushes is used as the driving force. We examined for a theory to control this robot by non-interference PID control, and clarified the effectiveness by experiments.
We developed a robotic hand that folds an origami form “Tadpole”. However, the robot, which simply replays a given trajectory, often fails in folding due to the fluctuation of origami paper behavior. In this paper, we propose a novel method to synthesize a desired trajectory and sensory feedback control laws for robots based on the statistical feature of direct teaching data demonstrated by a human. Hidden Markov Model (HMM) is used to model the statistical feature of human motion. Nominal desired trajectory is obtained by temporally normalizing and spatially averaging the teaching data. Sensory feedback control laws are then synthesized based on the output probability density function parameters of the HMM. Considering velocity variance and canonical correlation between velocity and force of the teaching data, important motion segments are extracted and feedback control laws are applied only for those segments. Experimental results showed that the success rate and folding quality of “Valley-fold” were improved by the proposed method. The proposed method enables robot motion teachers to simply perform direct teaching several times to transfer their skill, which is difficult to describe explicitly, to the robot.
It is expected that more and more robots will be introduced into public societies and homes. The problem in such a case will be the smooth mobility of each robot. Collision avoidance between robots or between a robot and other obstacles including human beings is important as well as reaching its target point in order to carry out its task. Human beings often give way to each other to cope with this collision avoidance problem in a decentralized manner in which they plan and move individually without mutually exchanging information. This paper demonstrates that the robot can produce giving way to each other behavior similar to human beings based on a decentralized approach. Spatiotemporal RRT, which is an extended version of Rapidly exploring Random Trees (RRT) that is a random sampling method for searching a large working space effectively, is used in order to produce this behavior. This paper also presents three typical types of environments that require giving way to each other and explains the result of numerical simulation with the proposed method.