The problem of computing volumetric shape from stereo is examined. It is argued that intermediate two-and-one-half-dimensional dense or wire-frame descriptions may not be always possible from stereo, especially when t...
详细信息
The problem of computing volumetric shape from stereo is examined. It is argued that intermediate two-and-one-half-dimensional dense or wire-frame descriptions may not be always possible from stereo, especially when there are curved surfaces in the scene, and that 3D volumetric descriptions of objects may have to be derived directly from stereo correspondences. Methods are then presented for recovering volumetric shape with linear straight homogeneous generalized cones (LSHGCs) and straight homogeneous generalized cones (SHGCs) as the shape models, using some invariant properties in their monocular and stereo projections. Experimental results on images of objects with curved surfaces are given.< >
A practical solution to the path planning problem for a six degree-of-freedom manipulator robot with arbitrary payload is described. The three phases of operation are explained. Phase one involves planning for the fir...
详细信息
A practical solution to the path planning problem for a six degree-of-freedom manipulator robot with arbitrary payload is described. The three phases of operation are explained. Phase one involves planning for the first three joints in a configuration space model using a distance transform. A real space model and heuristics are then used for the orientation planning of the wrist, gripper and payload. The execution of the path is monitored with sensor panels covering the robot so that safe adjustments to the path can be made if necessary.< >
In automated feature-based motion analysis of multiple frames, correspondence data are usually noisy and fragmented. A technique that gradually refines the initial noisy correspondence data and links fragments of a si...
详细信息
In automated feature-based motion analysis of multiple frames, correspondence data are usually noisy and fragmented. A technique that gradually refines the initial noisy correspondence data and links fragments of a single trajectory using feedback from 3D motion estimation is presented. First, 3D motion parameters are estimated using the initial correspondence data. Then, each noisy trajectory is partitioned into subsets of points, each of which conforms to the estimated motion. The best set is used as the input to the next motion estimation. This process is repeated, and the gaps in the refined correspondence data are filled by guidance from the predicted motion. Test results for a standard real image sequence are presented.< >
The effectiveness of the ellipsoid representation for geometrical reasoning is demonstrated in the context of robotics. Specifically, a robot collision-detection problem that consists of computing a quantity that refl...
详细信息
The effectiveness of the ellipsoid representation for geometrical reasoning is demonstrated in the context of robotics. Specifically, a robot collision-detection problem that consists of computing a quantity that reflects, as a function of the geometrical data, the amount of clearance between the robot and its environment is discussed. The method consists of two algorithms. The first computes the optimal ellipsoid surrounding a convex polyhedron. The second computes an analytic formula for the free margin about one ellipsoid with respect to another, as a standard eigenvalue problem. An efficient incremental version of the latter algorithm is proposed. This system has been implemented, and preliminary simulation results are provided.< >
An approach to the control of a robot manipulator in grasping a simple moving target with constant speed is presented. A layered neural network architecture-based controller has been developed. It can automatically le...
详细信息
An approach to the control of a robot manipulator in grasping a simple moving target with constant speed is presented. A layered neural network architecture-based controller has been developed. It can automatically learn visual motor coordination for the fast reaching movements required in grasping a moving target. A learning scheme known as two-phase learning is described for teaching the skill to the controller. Learning in the controller is achieved through a sequence of trial movements without the presence of a 'teacher'. Visual feedback showing the action of the controller is used to adapt it so as to reduce the error between the target and the robot's gripper.< >
The principle of a simple acoustic sensor system using three Polaroid transducers which can identify a plane, an edge, and a corner is discussed. Vector analysis of reflections involving one, two, and three mirror-lik...
详细信息
The principle of a simple acoustic sensor system using three Polaroid transducers which can identify a plane, an edge, and a corner is discussed. Vector analysis of reflections involving one, two, and three mirror-like orthogonal surfaces is presented. An algorithm to distinguish a corner from an edge using the triangular three-transceiver system is described, and computer simulated results are given.< >
暂无评论