Objects such as onions or other produce are often processed mechanically. This processing can include things such as chopping or peeling. To achieve this processing, objects often need to be loaded into a machine that mechanically processes the object. A challenge to loading objects into a processing machine is when the object is imperfectly spherical, such that its orientation going into the machine must be corrected in order to optimally process the object. Objects such as onions that are imperfectly spherical, and vary in size, present a challenge to loading into a processing machine. Manually orienting an object can be burdensome where large quantities of objects need to be oriented. As such, what is needed is an apparatus and method for accurately positioning an object on a surface.
The present disclosure is directed to an apparatus for positioning an object.
In one embodiment, the apparatus for positioning an object comprises four gripping elements, comprising two pairs of diametrically opposed gripping elements; four appendages, comprising two pairs of diametrically opposed appendages, operably connected to each of the four gripping elements; a vertical actuator, wherein the vertical actuator is operably connected to the four gripping elements, wherein the vertical actuator moves the four gripping elements vertically relative to the object; a horizontal actuator, wherein the horizontal actuator is operably connected to the four gripping elements, wherein the horizontal actuator moves each of the two pairs of diametrically opposed gripping elements from a first distance apart to a second distance apart, and wherein the second distance apart is sufficient to engage the object; and a rotational actuator, operably connected to the four appendages, wherein the rotational actuator rotates a first pair of the two pairs of diametrically opposed appendages in a first common direction about a first axis, wherein the rotational actuator rotates a second pair of the two pairs of diametrically opposed appendages in a second common direction about a second axis, wherein the second axis is perpendicular to the first axis.
In another embodiment, the apparatus for positioning an object comprises a frame; four arms arranged in radial symmetry around the frame, wherein the four arms comprise two pairs of diametrically opposed arms, wherein each arm comprises (i) a first end, wherein the first end is pivotly connected to the frame; (ii) a second end, wherein the arm extends vertically from the frame to the second end and wherein the second end is proximal to a surface; and (iii) an appendage, wherein the appendage is connected to the second end; a vertical actuator, wherein the vertical actuator is operably connected to the four arms, wherein the vertical actuator moves the four arms vertically toward the object, and wherein the vertical movement is determined by an object height; a horizontal actuator, wherein the horizontal actuator is operably connected to the four arms, wherein the horizontal actuator moves each of the two pairs of diametrically opposed arms from a first distance apart to a second distance apart, and wherein the second distance apart engages the object; and a rotational actuator, wherein the rotational actuator is operably connected to the appendage on each of the four arms, wherein the rotational actuator rotates the appendages on a first pair of the two pairs of diametrically opposed arms in a first common direction about a first axis, wherein the rotational actuator rotates the appendages on a second pair of the two pairs of diametrically opposed arms in a second common direction about a second axis, wherein the second axis is perpendicular to the first axis, each pair of the two pairs of diametrically opposed arms in a common direction to rotate the object.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a camera to capture at least one image of the object; an image processor configured to receive input from the camera, and a controller configured to direct movement of the rotational actuator, wherein the controller determines a rotational profile of the object and directs movement of the rotational actuator based on the rotational profile.
In any of the foregoing embodiments, the rotational profile of the apparatus for positioning an object is based on an object feature.
In any of the foregoing embodiments, the image processor comprises a pattern recognition algorithm.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a first height sensor operably connected to the vertical actuator, wherein the first height sensor measures a first height of the object above a surface, and wherein the first height of the object above the surface is the object height that determines the amount of vertical movement of the gripping elements by the vertical actuator.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a second vertical actuator, wherein the second vertical actuator is positioned below the four gripping elements, wherein the second vertical actuator moves the object vertically toward the four gripping elements, and wherein the vertical movement is determined by a second object height.
In any of the foregoing embodiments, the second vertical actuator of the apparatus for positioning an object comprises an end proximal to the object and an end distal to the object, wherein the end proximal to the object comprises fingers, and wherein the object rests on top of the fingers.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a light fixture.
In any of the foregoing embodiments, the light fixture of the apparatus for positioning an object is arranged in radial symmetry around the four gripping elements.
In any of the foregoing embodiments, the light fixture comprises a flat board surrounding the camera.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises four arms, each arm comprising a first end and a second end, wherein the first end is pivotly connected to a frame, wherein each of the four arms extends vertically from the frame to the second end, wherein the second end is proximal to the object, and wherein the second end of each arm comprises one of the four gripping elements.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a controller, wherein the controller is operably connected to the four gripping elements and the four appendages, wherein the controller controls movement of the vertical actuator, the horizontal actuator, and the rotational actuator.
In any of the foregoing embodiments, the apparatus for positioning an object further comprises a second vertical actuator, wherein the second vertical actuator is positioned below the four arms, wherein the second vertical actuator moves the object vertically toward the four arms, and wherein the vertical movement is determined by a second object height.
In any of the foregoing embodiments, the light fixture of the apparatus for positioning an object is arranged in radial symmetry around the four arms.
The present disclosure is also directed to a method for positioning an object.
In one embodiment, the method for positioning an object comprises measuring the object height with a height sensor; moving the four gripping elements vertically toward the object with the vertical actuator; moving a first pair of the two pairs of diametrically opposed gripping elements horizontally toward the object; gripping the object with the first pair of the two pairs of diametrically opposed gripping elements; lifting the object with the first pair of the two pairs of diametrically opposed gripping elements; rotating the object about the first axis with the first pair of the two pairs of diametrically opposed appendages; gripping the object with a second pair of the two pairs of diametrically opposed gripping elements; releasing of the object by the first pair of the two pairs of diametrically opposed gripping elements; and rotating the object about the second axis with the second pair of the two pairs of diametrically opposed appendages.
In another embodiment, the method for positioning an object comprises measuring the object height with a height sensor; moving the four arms vertically toward the object with the vertical actuator; moving the first pair of the two pairs of diametrically opposed arms horizontally toward the object with the horizontal actuator; gripping the object with the gripping elements on the first pair of the two pairs of diametrically opposed arms; lifting the object; rotating the object about the first axis with the rotational actuator operably connected to gripping elements on the first pair of the two pairs of diametrically opposed arms; gripping the object with the gripping elements on the second pair of the two pairs of diametrically opposed arms; releasing the object with the first pair of the two pairs of diametrically opposed arms; and rotating the object about the second axis with the rotational actuator operably connected to gripping elements on the second pair of the two pairs of diametrically opposed arms.
In another embodiment, the method for positioning an object further comprises the steps of measuring a second object height with a second height sensor, wherein the second object height is the height of the object below the surface; and moving the object vertically toward the four arms with a second vertical actuator, wherein the second vertical actuator is positioned below the four arms, and wherein the vertical movement of the second vertical actuator is determined by the second object height.
In any of the foregoing embodiments, the height sensor in the method for positioning an object senses a height of the object above the surface.
In any of the foregoing embodiments, the method for positioning an object further comprises the steps of measuring a second object height with a second height sensor, wherein the second object height is the height of the object below a surface; and moving the object vertically toward the four gripping elements with a second vertical actuator, wherein the second vertical actuator is positioned below the four gripping elements, and wherein the vertical movement of the second vertical actuator is determined by the second object height.
In any of the foregoing embodiments, the second vertical actuator of the method for positioning an object comprises an end proximal to the object and an end distal to the object and wherein the end proximal to the object comprises fingers.
In any of the foregoing embodiments, the method for positioning an object further comprises the steps of detecting the object with a camera, wherein the camera captures at least one image of the object; and creating the rotational profile of the object to determine the amount of rotation.
In any of the foregoing embodiments, the step of creating the rotational profile of the object in the method for positioning an object further is accomplished by a controller.
In any of the foregoing embodiments, the rotational profile of the step of creating the rotational profile of the object is based on the feature of the object.
In any of the foregoing embodiments, the method for positioning an object further comprises the step of exposing the object to a light.
In any of the foregoing embodiments, the object in the method for positioning an object comprises a spherical object.
In any of the foregoing embodiments, the object in the method for positioning an object comprises an onion.
In another embodiment, the method for positioning an object comprises exposing the object to a first sensor to measure a first dimension of the object; transmitting a first signal based on the first dimension from the first sensor to a controller, wherein the controller signals a movement to place the object in position for capturing a first image of the object; capturing the first image of the object thereby creating a second signal; transmitting the second signal to the controller, wherein the controller is programmed to create a first rotational profile along a first axis; and rotating the object along the first axis based on the first rotational profile.
In any of the foregoing embodiments, the method may further comprise the following steps after the object is rotated along the first axis: capturing a second image of the object thereby creating a third signal; transmitting the third signal to the controller, wherein the controller is programmed to create a second rotational profile along a second axis; and rotating the object along the second axis based on the second rotational profile.
In any of the foregoing embodiments, the method may further comprise, prior to transmitting a first signal based on the first dimension from the first sensor to a controller, exposing the object to a second sensor to measure a second dimension of the object, wherein the first signal is also based on the second dimension from the second sensor.
In any of the foregoing embodiments, the method may further comprise, prior to capturing the first image of the object, exposing the object to a light source.
In any of the foregoing embodiments, wherein the second signal is generated by a pattern recognition algorithm.
In any of the foregoing embodiments, wherein the first rotational profile is based on at least one object feature.
In any of the foregoing embodiments, the method may further comprise, prior to capturing the second image of the object, exposing the object to a light source.
In any of the foregoing embodiments, wherein the third signal is generated by a pattern recognition algorithm.
In any of the foregoing embodiments, wherein the second rotational profile is based on at least one object feature.
In any of the foregoing embodiments, wherein the second axis is perpendicular to the first axis.
In any of the foregoing embodiments, wherein the first sensor is operably connected to a vertical actuator, wherein the vertical actuator is operably connected to four gripping elements comprising two pairs of diametrically opposed gripping elements.
In any of the foregoing embodiments, wherein the second sensor is operably connected to a vertical actuator, wherein the vertical actuator is operably connected to four gripping elements comprising two pairs of diametrically opposed gripping elements.
In any of the foregoing embodiments, wherein the first sensor and the second sensor are operably connected a vertical actuator, wherein the vertical actuator is operably connected to four gripping elements comprising two pairs of diametrically opposed gripping elements.
In any of the foregoing embodiments, wherein the steps of capturing the first image of the object and capturing a second image of the object are performed by a camera, wherein the camera is operably connected to the image processor, and wherein the image processor is operably connected to the controller.
In any of the foregoing embodiments, wherein the steps of rotating the object along the first axis and rotating the object along the second axis are performed by a rotational actuator, wherein the rotational actuator is operably connected to four appendages, wherein the four appendages comprise two pairs of diametrically opposed appendages.
In any of the foregoing embodiments, wherein the two pairs of diametrically opposed appendages comprise a first pair of diametrically opposed appendages and a second pair of diametrically opposed appendages, wherein the first pair of diametrically opposed appendages rotates the object along the first axis, and wherein the second pair of diametrically opposed appendages rotates the object along the second axis.
The current invention is directed to an apparatus and method for positioning an object. The apparatus and method provide for positioning of objects of non-uniform size and shape to a desired position.
The present invention comprises four gripping elements capable of three different types of movement. The gripping elements may move vertically toward an object, horizontally toward the object, and rotate in a common direction so as to rotate the object. To provide this movement, the apparatus comprises a vertical actuator, horizontal actuator, and rotational actuator. The apparatus may further comprise a camera to capture at least one image of the object and an image processor configured to receive input from the camera. The image processor may comprise a pattern recognition algorithm that extracts one or more features of the object and registers the one or more features along an orientation axis. The apparatus may further comprise a controller configured to determine a rotational profile of the object features and direct movement of the rotational actuator to rotate the object to the desired position. The apparatus may further comprise a one or more height sensors operably connected to the vertical actuator or a second vertical actuator, measuring the height of the object above or below a surface. The apparatus may further comprise a light fixture to shine light on the object. The apparatus may further comprise four arms extending from a frame, with gripping elements at the ends of the arms that contact the object to be positioned.
The schematics depicted in
In some embodiments, the positions of the first height sensor 20a and the second height sensor 20b determine whether the second vertical actuator 50 engages. By way of example but not limitation, if the object position measured by the first height sensor 20a is below a threshold, the second vertical actuator 50 is engaged to lift the object. If the object position measured by the first height sensor 20a is above a threshold, the second vertical actuator 50 is not engaged and does not lift the object.
In an embodiment, the following steps are carried out: An onion is in position 1 on a surface and it is desired for the object to be in position 2 on the surface. The height of the onion is measured with a height sensor. Four gripping elements move vertically toward the onion. A first pair of diametrically opposed gripping elements grip the onion and lift the onion. A camera captures a first image of the onion. A pattern recognition algorithm is applied to the first image of the onion and registers the X and Y coordinates of one or more features of the onion such as root, node, stem, neck, and meridian an lines, using the dead center of the onion in the first image as the origin of the axes. The controller receives the X and Y coordinates of the one or more features of the onion, generates a first rotational profile that represents the angle that the onion must be rotated to a desired position on a first axis, and converts the first rotational profile to motor counts and commands a first rotational actuator to rotate along the first axis. Two diametrically opposed appendages on the first pair of diametrically opposed gripping elements rotate the onion about a first axis. A second pair of diametrically opposed gripping elements grip the onion. The first pair of diametrically opposed gripping elements release the onion. The camera captures a second image of the onion. The pattern recognition algorithm is run again and the controller generates a second rotational profile that represents the angle that the onion must be rotated to a desired position on a first axis, and converts the second rotational profile to motor counts and commands a second rotational actuator to rotate along the second axis. Two diametrically opposed appendages on the second pair of diametrically opposed gripping elements rotate the onion about a second axis. The vertical actuator lowers the onion to the surface. Optionally, after rotation on each axis, the camera may capture additional images of the onion and additional rotational profiles along each axis may be generated by the controller to determine if the onion is in the desired position on each axis. Optionally, at any step of the preceding method, the controller may determine that there is not enough time to rotate the onion and may signal the vertical actuator to lower the onion to the surface.
Gripping Elements. The apparatus comprises four gripping elements. The four gripping elements comprise two pairs of diametrically opposed gripping elements. In an embodiment depicted in
Appendages. In an embodiment, the apparatus comprises four appendages, comprising two pairs of diametrically opposed appendages. In the embodiments depicted in
Vertical Actuator. The apparatus comprises a vertical actuator that moves the four gripping elements vertically toward the object. The vertical actuator is operably connected to four gripping elements. In the embodiment of the invention where each gripping element 30a, 30b is connected to an arm 60, as depicted in
Horizontal Actuator. The apparatus comprises a horizontal actuator operably connected to the four gripping elements. The horizontal actuator functions to move each of the two pairs of diametrically opposed gripping elements to clamp onto the object. By way of example but not limitation, as depicted in
Rotational Actuator. The rotation on a pair of diametrically opposed gripping elements functions to rotate the object in a common direction, meaning that the rotation of the two diametrically opposed gripping elements is opposite, one gripping element rotating in a clockwise fashion and the corresponding diametrically opposed gripping element rotating in a counterclockwise fashion, to achieve rotation of the object in a common direction. By way of example but not limitation, the appendages 35a of the two diametrically opposed gripping elements 30a depicted in
Camera. In some embodiments, the apparatus comprises a camera. In one embodiment, depicted in
Image Processor. The apparatus may further comprise an image processor. The image processor receives an image of the object from the camera. The image processor may pre-process images received from the camera, such as removing blurs and shadows. In some embodiments, the image processor comprises a pattern recognition algorithm. The pattern recognition algorithm extracts a feature of the object from the image, and registers the feature location along an orientation axis of the image. In some embodiments, the pattern recognition algorithm registers a feature at X and Y coordinates of a two dimensional image of the object, with the dead center of the object in the two dimensional image being the origin of the X and Y axes. Any one or more recurring features present on the objects to be positioned can be used in the image processor. By way of example but not limitation, different object features that could be used in the image processor include roots (i.e., the bottom of the object), meridian lines (i.e., lines traversing the object between the top and bottom), nodes (i.e., the top of the object), stems, necks, and shape. In some embodiments, the pattern recognition algorithm comprises a neural network-based feature detector comprising a database of tagged features for deep learning and different object topologies. In some embodiments, the pattern recognition algorithm comprises custom software written in the C programming language. In some embodiments, the image processor carries out any or all of the functions of network communications, image capture, image processing, and output filtering.
Controller. The apparatus may further comprise a controller. The controller functions to translate input from components of the apparatus into movement to achieve desired re-positioning of the object. By way of example but not limitation, the controller may control any or all of the following: amount of vertical movement by the vertical actuator; amount of horizontal movement by the horizontal actuator; amount of rotation by the rotational actuators. In some embodiments, the controller functions to create the rotational profile of the object based on input from the image processor.
By way of example but not limitation, in controlling vertical movement, the controller may translate the height of the object measured by a height sensor to a desired vertical movement, then command the vertical actuator to move the amount of desired vertical movement. In some embodiments, the controller may use extra modifying terms to artificially optimize the geometry of the object to achieve desired movement. The controller may also include object-to-surface transition code tuning, which can (1) account for non-spherical object shapes; and (2) lower the gripping elements an optimal distance to achieve release of the object without the object becoming misaligned. The controller may also synchronize vertical movement in embodiments where there are two vertical actuators. The controller may synchronize horizontal movement of the horizontal actuator.
In controlling rotational movement, the controller may create one or more rotational profiles. Such rotational profiles comprises the one or more angles that the object must be rotated to achieve the desired object orientation. In some embodiments, the controller creates the rotational profile by translating feature location along an orientation axis provided by the image processor into an angle that the object must be rotated to move the features into a desired position.
In an embodiment, the controller generates a first rotational profile to determine the angle that the object must be rotated to achieve the desired object orientation along a first axis, and the controller generates a second rotational profile to determine the angle that the object must be rotated to achieve the desired object orientation along a second axis. The second rotational profile may be based on feature locations from one or more images taken before or after rotation along the first axis.
In some embodiments, a first rotational profile is created when the controller receives X and Y coordinates of an object feature from the image processor and converts the X and Y coordinates of the feature to a rotation angle along a first axis to achieve the desired position on the first axis by using an approximation of object diameter. By way of example but not limitation, the approximation of object diameter may be based on the height of the object measured by a height sensor. Once the controller has converted X and Y coordinates to rotation angles along the first axis, the controller may either convert the angle to motor counts and command a first rotational actuator to rotate along the first axis; or the controller may switch to control of a second rotational actuator. In some embodiments, the controller may create a second rotational profile for the first axis by receiving a second set of X and Y coordinates of an object feature and converting the second set of X and Y coordinates to a second rotation angle along a first axis. In some embodiments, the controller generates a rotational profile along a second axis by receiving additional X and Y coordinates of an object feature from the image processor and converting the additional set to a rotation angle along a second axis using the same process as conversion of X and Y coordinates to an angle for rotation along the first axis. The controller may either convert the angle to motor counts and command a second rotational actuator to rotate along the second axis; or the controller may switch to control of the vertical actuator and lower the object to the surface. In some embodiments, the controller may create a second rotational profile for the second axis by receiving a second set of X and Y coordinates of an object feature and converting the second set of X and Y coordinates to a second rotation angle along a second axis. The controller decision to command a rotational actuator to rotate along an axis may depend on, by way of example but not limitation, variables such as a set time remaining for rotation or the amount of correction needed along the axis to achieve the desired object position.
In some embodiments, the controller determines the time remaining for rotation of the object and commands the vertical actuator to lower the object to the surface if there is insufficient time. By way of example but not limitation, the controller may make this determination by a set value of time permitted for each object; how much time each type of action requires; or how fast the surface is moving. In some embodiments, the controller may determine that there is not enough time to perform rotation to achieve the desired position, the object is set down with rotation only along one axis or no rotation at all.
In some embodiments, the controller may account for errors in the image processor. In some embodiments, the controller creates the rotational profile by translating two-dimensional feature location into an object angle, wherein the geometry of the object is experimentally optimized with extra modifying terms and wherein the controller accounts for errors due to three-dimensional to two-dimensional information loss and changes in trigonometric function sensitivity at large angles. The controller may also include rotation logic tuning for speed; calculating for error by two-dimensional distance or by angle; or mapping the location of features on the object using a polar or a Cartesian approach. In some embodiments, the hardware component of the controller is B&R Industrial Automation's X20 Series Controller.
Height Sensor. In some embodiments, the apparatus comprises one or more height sensors. In one embodiment, as depicted in
Light Fixture. The apparatus may further comprise a light fixture. By way of example but not limitation, as depicted in
Arms. The apparatus may further comprise four arms extending vertically toward the surface. By way of example but not limitation, as depicted in
Frame. In some embodiments, the apparatus comprise a frame. The frame may connect all components of the apparatus or may just connect select components of the apparatus. The embodiment of the invention depicted in
Surface. The apparatus functions above a surface. By way of example but not limitation, as depicted in
Object. The apparatus may position a variety of objects. By way of example but not limitation, as depicted in
The present application is a continuation of PCT/US19/62250, filed Nov. 19, 2019, which claims the benefit of U.S. Provisional Application No. 62/769,443, filed on Nov. 19, 2018 entitled, “Apparatus and Method for Positioning an Object”, the entirety of which are incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
62769443 | Nov 2018 | US |
Number | Date | Country | |
---|---|---|---|
Parent | PCT/US19/62250 | Nov 2019 | US |
Child | 17325084 | US |