Abstract: In a method and system for generating teaching data for robots, a plurality of virtual target points are disposed in the vicinity of each of target points allowable on design, where a tool center point of a robot is to be positioned. The virtual target points and the original target point are combined into "target points taken in a broad sense". Each of the target points taken in the broad sense is given as a point which has been confirmed as a tool-attainable point on an off-line teaching apparatus. When a necessary correction is made on-line to teaching data generated by the off-line teaching apparatus, including as data the target points taken in the broad sense, an operator can select a suitable point from the target points taken in the broad sense and carry out a process of correcting the teaching data, regarding the suitable point as being a starting point.

Abstract: A multi-axis robot and controller therefor which is responsive to the input of target position information for each of plural axes and outputs position commands to motors for each axis of the robot. All of the axes are considered together and are assumed to move according to the same velocity profile, the fastest velocity is identified and used as the basis for assigning a relative load factor to each axis. An actual velocity command for movement of each axis according to the input target position is calculated on the basis of the preceeding position command, its load factor, velocity profile and interpolated velocity value. The position command is calculated on the basis of the velocity command. The controlled axes may include additional outside axes and the load factor may be variable.

Abstract: A knowledge-based system (10) which combines case-based reasoning, heuristic search and deductive rule application. The resulting inference engine is sensitive to the context of problem solving. The system (10) includes a heuristic searcher (22), a case memory (14), a rule memory (18), a rule applier (26) and a case matcher (24). The rule applier (26) uses stored rules to elaborate on new cases by deriving new features so that it will be closer to selected old cases. The case matcher (24) detects how close selected cases are to the new case and generates a score for the match. The heuristic searcher (22) maintains a plurality of elaborated cases and determines the goodness of each elaborated case, the goodness being a measure of the match between the elaborated case and the selected old case.

Abstract: A signal discrimination device using a neural network for discriminating input signals such as radar reception signals includes an adaptive code generator means for generating codes for representing the discrimination categories. The distances between the codes for closely related categories are smaller than the distances between the codes for remotely related categories. During the learning stage, the neural network is trained to output the codes for respective inputs. The discrimination result judgment means determines the categories by comparing the outputs of the neural network and the codes for the respective categories.