No abstract

No abstract

A reinforced concrete box culvert (RCBC) is a box culvert constructed of reinforced concrete. Because the sudden collapse of an RCBC could result in a serious accident, continuous management and maintenance work of the RCBC should be performed. In this paper, the RCBC chipping robot that is developed as an alternative to skilled construction workers for carrying out the concretechipping work, one of the RCBC maintenance process, is introduced, and the force-regulated impact control for improving quality of the chipping work is described. In order to implement the proposed algorithm, reaction forces when the robot strikes concrete with chipping tool are collected, and classification to identify the striking object and its hardness is carried out based on the obtained reaction forces. In addition, a variable motion control algorithm is realized by setting the dynamic characteristics of the robot with respect to the reaction force according to striking object and its stat. To verify the effectiveness of the proposed algorithm, chipping work with an RCBC chipping robot is conducted in a simulation environment consisting of reinforced concrete.

In this paper, a method for a biped robot to run with a velocity change by selecting its foot landing position appropriately is proposed. The zero moment point (ZMP) is used as the criteria to identify the stability of the biped robot. The ZMP is estimated by the position of the center of mass (COM) and the acceleration of the robot. This means that by setting an appropriate ZMP for the robot, the acceleration of the robot can be controlled. In this paper, ZMP is not only used for stability criteria, but also used to change the velocity in running. The ZMP is designed to reach the desired velocity of the robot, and the trajectory to change the running velocity is generated made the assumption that the ZMP is the foot landing position. By using the method, the biped robot can run stably while changing its velocity. The performance of the proposed method was verified by computer simulations.

This paper proposes the redundancy trajectory generation method for biped robots. First of all, the position and orientation variables of a robot's foot and body are chosen as the reference factors. And the augmented Jacobian method is used at the acceleration level with the differential equation of reference factors and ZMP equation to guarantee the minimum motion factor and the stability condition. The virtual spring-damper system is reflected as an additional task at the null-space component to make a symmetrical walking motion. To prove the effectiveness of the proposed the redundancy trajectory generation method, computer simulations are carried out.Index Terms--biped robot, redundancy system, virtual spring-damper system, augmented Jacobian method.