A novel approach based on the use of force sensors for motion in contact with uncertainty in task planning is presented. A neural network monitors the force signals measured by a sensor mounted in the robot wrist. This network is able to learn without need of a teacher the different contact states of the system. The method is intended to work properly in complex real-world situations, for which a geometric analytical model may not be feasible, or too difficult. In this paper the authors study the two-dimensional peg-in-hole problem and a real example of a complex insertion task in a flexible manufacturing system.

A wall-climbing robot intended for inspection in nuclear power plants has been developed. The robot, small in size and modular, is made up of pneumatic components exclusively. Vacuum suction cups are used for sticking the robot to walls to be climbed. The robot motion, made up of both elementary translations and rotations, are accomplished by means of flexible pneumatic cylinders of rubber, which constitute the body of the mobile robot. A personal computer is used to command robot motion and a microprocessor controls the electronic valves of all pneumatic components. The paper presents the robotic system conceptual design and gives a detailed description of the mobile robot and its motion.>

The productivity of present day industrial robots may be hampered by phenomena such as vibrations and deformations which appear whenever they are made to move at high speeds or to carry heavy payloads. To address this problem, many researchers have studied the flexibility of robot arms. Most works, however, rely on simple one or two-link flexible arms made with extremely flexible beams which operate at very slow speeds. Thus their results may prove difficult to apply to more realistic situations. The CEIT flexible robot is a lightweight three degree of freedom flexible arm which has been built to bridge this gap between flexible and industrial robots. The structural behavior of this robot is investigated through experimental modal analysis. The procedure and results of the analysis are described in this paper.>

By combining the simplicity of the sphere and the power of the motion of hierarchy of detail, a novel model is proposed with applications to collision avoidance and detection in 3D. The model is based on a double spherical representation for solid bodies. First, each element making up the robot and the obstacles is approximated by a set of exterior spheres which are automatically defined. Second, another set composed of interior spheres is generated. These representations define a hierarchy, since they can be redefined as many times as necessary; starting with two spheres per element, the approximation may be improved until it contains hundreds of spheres. Moreover, they converge to a zero-error representation. The proposed models leads to a simple treatment for the problem of collision detection, and it is further applied to collision-free path planning for manipulators in 3D.>

A trajectory planner for a one-flexible-link manipulator is presented. The approach is based on the solution to the inverse dynamic problem for flexible manipulators in the frequency domain. The problem is mathematically formulated as an optimization of the settling time subject to maximum torque and high frequency content constraints. A cubic-rectangular acceleration profile is proposed as a generic input to be used in the optimization process, and the general optimization procedure is established. The application of the approach to a particular case shows the operativeness of the proposed methodology and the possibility of an intelligent election of trajectories for flexible manipulators.>