J. Lückel, W. Moritz, W. Kuhlbusch, F. Scharfeld, St. Toepper, P. Maißer, H. Freudenberg, E. Kallenbach, J. Zentner, E. Saffert, Universität Paderborn, Germany
The TRIPLANAR is a novel parallel robot with six degrees of freedom for highly precise measurement and process tasks. Its prototype was developed in just 12 months by means of the mechatronic composition. This design methodology comprises analysis and synthesis of the entire system, including mechanics, actuator system, and information processing in the computer before the constructive design is started. The information processing was designed by means of a simplified model and tested with a complex model. An experimental modal analysis of the real system served to identify the parameters of the complex model. In order to increase the limited motive forces of the TRIPLANAR novel drives, so-called …
P. Lucidarme, P. Rongier, A. Liégeois, LIRMM, France
This paper presents the implementation of an experimental setup to study the behavior of a group of purely reactive mobile robots subject to serious unrecoverable failures: some of them have lost mobility, others have lost infrared vision. To survive, each robot must reach a supply center. The eventually degraded mission uses complementary capabilities of the two types of robots by letting a blind robot, meeting by chance an invalid one, carry it. The latter is able to "see" a supply center within a wide range. The hardware and software of the experimental devices are described, using very simple and low-cost components. Experiments are run with the robots in different initial positions. The results are compared to those obtained by many computer simulations and by a Markovian process model, both are also described.
A New Method for Fully Autonomous Calibration of Parallel Manipulators Using a Constraint Link
Jeha Ryu, Abdul Rauf, Kwangju Inst. of Science and Technology, Korea
Kinematic calibration is a process whereby the actual values of geometric parameters are estimated so as to minimize the error in absolute positioning. Fully autonomous methods of calibration are cost effective fast and easy to implement. This paper presents a new method for fully autonomous calibration that can be applied to 6 degree-of-freedom fully parallel manipulators. The proposed method uses a link of fixed length having spherical joints at both of it ends to restrain the motion of the end-effector over spherical surface with arbitrary orientations. The end-effector therefore has 5 degrees of freedom. However the measurement data is collected from 6 actuator sensors. This redundancy of sensory information is utilized for calibrating the geometric parameters without using any extra sensing device …
12:10―12:30
Adaptive Relocation of Environment-Attached Storage Devices for Effective Knowledge-Sharing among Multiple Robots
We propose a method of effective knowledge-sharing among multiple robots by using storage devices which can be attached into environment. If robots store knowledge about environment to the devices robot can use the stored knowledge and redcuce cost for achieving tasks. Since memory size of such storage device is limited robots cannot store all of required knowledge. Therefore the robots store knowledge according to its usefulness defined by reducible cost per time. A memory unit storing useless knowledge is removed from the device and added to the device that requires more useful knowledge. Then the robots can store useful knowledge over the whole environment and the memory units are flexibly relocated …
Kinematics of the Translational 3-URC Mechanism
Raffaele Di Gregorio, Università di Ferrara, Italy
A new three-dof spatial parallel mechanism, named translational 3-URC, is presented. The translational 3-URC makes the platform translate with respect to the base. The new mechanism belongs to a set of mechanisms with topologically equal architectures, called 3-URC, which contains another mechanism behaving as a spherical parallel wrist. The translational 3-URC position and mobility analyses are addressed and solved in closed form and the singularity conditions are written in explicit form and interpreted.
12:30―12:50
Collision Detection of Manipulator Based on Adaptive Control Law
Taishi Matsumoto, Kazuhiro Kosuge, Tohoku University, Japan
In this paper,we propose a collision detection method of a manipulator based on the nonlinear adaptive control law proposed by Slotine and Li. The collision of a manipulator with its environment is detected by the difference bet een the actual input torques to the manipulator and the reference input torques calculated based on the manipulator dynamics.An adaptive control scheme is employed for the manipulator control and the parameter estimation of the manipulator.The reference input torques are calculated using the estimated manipulator parameters. The proposed collision detection scheme is applied to an industrial manipulator and the experimental results illustrate the validity of the proposed scheme.
An Experimental Validation of CaPaMan as Earthquake Simulator
Chiara Lanni, Fabio Pugliese, Marco Ceccarelli, Università di Cassino, Italy
