In this paper, we propose a repeated impulse force generator by an active pantograph with a rubber as an energy storage. The proposed impulse force generator consists of a rubber which has high energy storage capacity per unit mass, only one motor and some generic mechatronics elements. A robot based on this impulse force generator has the repeated jumping capability to jump on the desk. The most important feature of this impulse force generator is that it can hold reaction forces of the rubber and release the stored elastic energy without additional trigger mechanisms. From numerical simulation based on a simple mechanical model, we show that this mechanism can make a large deformation in a very short time, which is utilized as impulse force generation. We show that the robot with a weight of 0.127[kg] achieves a jump up to the height of about 1.05[m], which means that the robot has the ability to jump onto a normal desk easily. We also show that the robot has a good payload characteristic, i.e., its jumping ability is not declined against payload variation by selecting an appropriate rubber with the optimal spring constant.
We have developed a new power assist arm for handling heavy loads using large pneumatic muscles. We chose to use the pneumatic muscles from a viewpoint of safety, but large actuators were required for heavy loads. As a result, it was expected that control becomes difficult due to the increased air flow requirement. Therefore, to optimize the hardware, we used silicone rubber filled pneumatic muscles in order to reduce a flow of air. And we increase the total cross-sectional area of a flow. On the control side, we developed an original pneumatic muscle model and the torque control based on it. Furthermore, we developed an impedance control system and successfully implemented it as part of a working power assist arm for handling heavy loads.
This paper presents a method for trajectory planning of multi-body systems. First, trajectory planning of a multi-body system is formulated as a constraint solving problem on a set of variables expressing the motion of the multi-body system over a finite time interval. Constraints express kinematic and dynamical conditions, range limits as well as task achievements, and they can be assigned different priority levels. The prioritized constraint-solving problem is treated under the framework of lexicographical goal programming. The local optimality of the problem is characterized in terms of Pareto efficiency condition. Based on this observation, an algorithm that iteratively updates the variables towards a locally optimal solution is derived. The proposed trajectory planning method is applied to target-reaching tasks of 3-DOF and 6-DOF robotic manipulators.
This paper discusses the analogy between human and robot under the novel concepts of “agonist-antagonist muscle-pairs ratio (A-A ratio)” and “agonist-antagonist muscle-pairs activity (A-A activity).” These concepts are similar to the equilibrium point hypothesis, but, we make clear linearity between the A-A ratio and the equilibrium point and clarify linearity between the A-A activity and the joint stiffness. Here we represent a new interface, constructed by using a simple translation from EMG to the inputs of pneumatic artificial muscles, to connect a biological system with a robot system, and demonstrate how to treat the two systems by using our concepts.
This paper discusses points of issue on static analysis of indeterminate contact forces raised by Omata. In summary, 1) Rigid-body-based analysis of static contact forces may have an advantage over elastic-contact-based analysis, which is physically more accurate, from an engineering viewpoint. 2) The elastic-contact-based method for contact force analysis by Rimon et al. is applicable only when preload contact forces are known and therefore it cannot always substitute for the conventional rigid-body model with Coulomb friction. 3) Virtual slidings and their combinations used in rigid-body-based analysis by the authors correspond to actual physical phenomena of contact forces.
The purpose of this study is to develop teaching materials for “Roboelecom education” in the technology education of the junior high school. The authors developed the educational compact computer, the sensor robot, the sensor light and the textbook. Then we organized the experiment. As a result, we confirmed that “the students learn quite easily to create simple programs”. Therefore we judged that “the programming learning using the educational compact computer is possible”. Moreover we confirmed that “these teaching materials support “Roboelecom education”.
This paper presents the introductory education to the robotics for a university new student. Introductory programs and course materials which utilized robot manufacturing are proposed. This program outlines a robot's structure, component technology, and a robot system. University new students study the robotic course by manufacturing the robot using the course materials. As a result of the evaluation experiments, proposed introductory program and depeloped robot manufacturing was verified that it was suitable for introductory education.