Kabushiki Kaisha Toyota Chuo Kenkyusho

Toyota Jidosha Kabushiki Kaisha

Kabushiki Kaisha Toyota

Masaru Nakano

Toshitaka Kuno

Yasuo Ishiguro

Yoshito Kato

Mitsuo Koide

Hiroshi Moribe

Chisao Hayashi

Minoru Tanaka

Kazutoshi Sukigara

Yoshiaki Ito

Osamu Shiroshita

Toshiaki Ikeda

Yoshizumi Itou

Christoph Hermann Roser

Toshio Kato

Fumiaki Takeda

A method of determining the level of the effect of each system entity on the overall system performance, wherein the status of each entity at each one of discrete points of time is determined as the corresponding one of an active and an inactive status; the length of the active period in which each entity is situated without interruption in the active status is determined as an active time; and, on the basis of the relationship in magnitude between the system entities with respect to the determined active time, the level at which each entity affects the performance of the system is determined.

Method of determining level of effect of system entity on system performance, by use of active time of same entity

A basic image of objects is extracted from a two-dimensional image of objects. Geometrical elements of the objects are extracted from the extracted basic image. The objects to be recognized are identified by searching a combination of the geometrical elements which match a geometrical model and then utilizing candidate position/orientation of the objects to be recognized, said candidate position/orientation being determined from a relationship in relative position between the combination of geometrical elements and the geometrical model. Mesh cells fixed to the geometrical model are mapped on the basic image based on the candidate position/orientation. In addition, verification is made as to whether an image of the geometrical model mapped by the candidate position/orientation is accurately matched with an image of one of the objects to be recognized, through a neural network to which values got from the basic image included in the individual mesh cells are to be applied as input values. Combination weight factors employed in the neural network are learned according to the verified results. It is also possible to recognize the multi-purpose objects according to how to learn the combination weight factors.

Apparatus for visual recognition

An apparatus for detecting the collision of objects including a moving object includes a face information memory for storing face information describing the surfaces of each object; hierarchical sphere model generator for modeling the surfaces of each object by hierarchically covering the surfaces with spheres having various radii on the basis of the face information; sphere model memory for storing the positions of the modeled spheres generated; sphere position updater for updating the position of each modeled sphere having various radii with respect to the moving object and outputting the updated position; and sphere collision computer for executing procedures in which mutually colliding spheres are detected on the basis of the positions of the modeled spheres concerning two of the objects generated at an arbitrary time and stored in the sphere model memory and colliding spheres among spheres belonging to the mutually colliding spheres and having a smaller radius are further detected with respect to only the spheres determined to collide with each other, and for determining the collision of the two objects with each other when finally mutually colliding spheres having the smallest radius are present. The procedures are executed hierarchically and consecutively starting with spheres having a large radius toward spheres having a smaller radius.

Apparatus for detecting the collision of moving objects

This invention is a real-time tracking controller for a standard-path robot that detects a target point in advance, permitting higher-speed work. The robot uses both sensed data and previously taught data to accurately execute its task. Accurate control is obtained using target points corresponding to time intervals. Thus, even if the speed or path is changed during operation, accurate tracing is continuously executed.

Real-time robot control system tracking based on a standard path

In an image processor system, a processor includes a first two-dimensional look-up table for receiving first and second digital data as an address, a second two-dimensional look-up table for receiving an output digital data of the first two-dimensional look-up table and a third digital signal as an address, and a latch circuit for temporarily storing the output digital data of the second look-up table. The processor is flexible and the functions thereof can be varied by changing the contents of the look-up tables.

Image processor system having multifunction look-up table units

A tracking control robot controls its arm end, which holds a tool, to track a taught path so that a reaction force between a workpiece and the tool remains constant. An ideal state for the arm end is determined from the present state of the arm end and the difference between the actual reaction force and a predetermined target force. Then, this ideal state is compared with a reference state of the arm end on the taught path to obtain a target position and a target attitude of the arm end. Since a target position is always determined independently of the last target position, no position error due to time delay affects a new target position even when the next target position is generated before the tool reaches the last target position. Thus, the robot is quick, responsive, and accurate when tracking at a constant operation speed.

Robot with controlled tool tracking displacement

A system is provided for controling a robot operating in association with a rotary table, including a robot, a rotary table, and a control circuit. The control circuit carries out a step of composing and storing teaching data as values on a rotary table orthogonal coordinate system (UVW axes), the rotational shaft of the rotary table being the vertical (W) axis, a step of calculating the UVW coordinates of a plurality of interpolation points by carrying out path interpolation along a line and/or curve defined by the stored teaching data, a step of calculating the degree of rotation of the rotary table which is required to bring each of the interpolation points into the U-W plane, a step of computing the UVW coordinates of the positions that the interpolation points occupy as they pass through the U-W plane, a step of transforming the data obtained by the computing step from coordinates in the UVW coordinate system to coordinates in a robot assigned coordinate system (XYZ axes) or orientation positions of the joint axes of the robot, and operating the robot in accordance with the information thus derived.

System for controlling a robot in association with a rotary table

A control system for an industrial robot having a hand which traces a pre-stored standard course defining its position and posture and which has a foresight function. The hand is provided with a tool and a work shape sensor, where the tool and the work shape sensor have a known spatial relationship. The foresight function is realized by a control system comprising temporary storing means for temporarily storing a future position and/or posture data of the tool calculated from a sensed position and/or posture data of the present sensor position. The stored data is output after a delayed time interval when it is compared with a pre-stored standard data. When the difference is small, the data is used to control the future position and/or posture of the hand. When the difference is great, an abnormality process is started.

Growing community of inventors

Nagoya, Japan

Masaru Nakano