ES2253135B2 - ARTIFICIAL VISION SYSTEM TO COLLECT SMALL FRUITS CULTIVATED IN ROWS. - Google Patents

Abstract

Artificial vision system to collect small fruits grown in rows. This invention consists of a vision system that allows the precise guidance of a robotic device (6) for the apprehension, cutting and deposition of the fruit in the desired place. The system accurately designates the exact 3-D point to which to direct the terminal mechanism (7) that will grab the peduncle of a ripe fruit and cut it over the predetermined distance, distinguishing it from the peduncles of other nearby fruits that are not due cut. The basic vision system is composed of: two color cameras (1), an array of low-cost spot laser diodes (spot type) (2), mounted on a platform (3) adjustable in azimuth and elevation, which is use to orient the matrix to the desired angle; and an additional laser diode (4), capable of projecting a flat or fan optical beam.

Description

Artificial vision system to collect Small fruits grown in rows.

Technical sector

The invention is part of the sector of automated systems for use in agriculture, and specifically in the of robotic systems, based on artificial vision, for tasks of handling and harvesting in horticulture.

State of the art

The interest of the production sector is known agrarian for having automated, automated or robotic systems semi-automated, fruit collection for different cultivation modalities, including tree fruits, fruits grown in soil or fruits grown high, as in the hydroponic crops This invention relates to the collection of small delicate fruits grown high, such as Strawberries in hydroponic greenhouse crops. In this kind of crops, fruit plants grow in rows, planted on a physical support, such as plastic gutters, which disposed longitudinally, at a variable height above the floor of the  greenhouse. The fruits grow and ripen with the hanging peduncle out of the culture support. In some facilities, the rows can be raised or lowered to different heights, individually or in groups, to facilitate access to plants in cultivation and harvesting tasks.

The general technological problem to solve is the of the conception of an autonomous vehicle, equipped with the systems of vision and manipulation necessary to carry out a collection careful of delicate fruits, with minimal human assistance. For To be viable, the system must satisfy at least the following three requirements:

First, it must be able to collect the fruits with extreme delicacy, to ensure maximum durability and quality. In particular, you should collect without apprehending the fruit, and deposit it in the storage container without subjecting it to any process that involves knocks or friction.

Second, to ensure efficiency and profitability of the work, the system must be able to collect all ripe fruits and not only those that are easily Identifiable for being isolated. Therefore, you must be able to recognize and isolate fruits that grow in clusters, being understand by cluster, for these purposes, the spatial concentration of Various fruits It must also be able to solve the problem of occlusions, especially those produced by the partial concealment of ripe fruits harvested by others not mature.

Third, for the same efficiency reason it has to be able to collect the fruits at a reasonable speed high. In this context, speed should be understood reasonably high one that is limited only by capacity of movement of the selected robotic device and not by limitations on acquisition, operation and processing time of fruit detection and location systems. This circumstance imposes explicit requirements on the detection system and location.

Well, there are no practical devices capable of satisfying these three requirements simultaneously in the desired grade The main limitation is, so far, in the vision system itself, understanding as such the set of means and algorithms whose mission is to detect and locate, with the required precision, the position of the ripe fruit and, singularly, of his peduncle in provisions of fruits in cluster. The solutions for fruits or vegetables that always grow in isolation (cucumber, eggplant, etc.) are unsuitable for collection of delicate fruits that can grow together.

The vision systems proposed for collection applications are preferably based on use of a camera, whose output is processed to detect by color, curvature or shape the presence of the fruit and make a first location in the plane of the 2-D image. For determine the third coordinate (depth or distance of the fruit to the vision system), the 2-D information of the camera is complemented with information from the third coordinate range (distance), which supplies a measuring device for distance. This can be a device. ultrasonic emitter-receiver or a couple of laser-type devices (transmitter and receiver) operating in the infrared or visible. This is the foundation, for example of the systems proposed in patents JP8112021, JP2004180554, JP5174131, JP5168334, JP5168333. In most cases, the distance determination or distance map calculation it is done by triangulation, estimating the delay or the difference phase between the signal emitted by the sending device and the Received at the receiver. Thus, in JP5168334 the radiation emitted by the emitter is deflected with a controlled mirror mechanically, which focuses the radiation towards the area to be explored, where it is reflected to be captured by the receiving device or eye electronic. Known layout geometry, for each deflection angle, a simple trigonometric triangulation It allows to calculate the distance of the reflection point. Unfortunately, these systems have a spatial resolution limited, probably enough for some applications but insufficient to discriminate with the required accuracy the position of a small fruit in a cluster and especially the position 3-D peduncle. Any increase in resolution possible demands a slower deflection process and consequently a lower operating speed in the collection.

In these systems, the camera and / or the meter distance can be fixed on the vehicle or mounted on the robotic arm end picker.

In other proposals specifically designed for harvesting medium or large sized fruits that grow isolated (see for example JP2001095348), the vision system is constituted by two cameras that operate sequentially, a fixed one that  analyze the whole scene globally and locate the fruit in the 2-D plane and another, mounted on the end of the manipulator, which guides the arm as it approaches its objective.

Another theoretically possible solution to solve the problem could be thought to be the long known of the stereoscopic vision, consisting of having two cameras with overlapping vision and with slightly different perspectives of the scene. The unknown 3-D position of a given point in the scene can be determined from their positions 2-D in the images captured by the cameras (projections). These systems do not seem to have at the moment use practical in the type of collection of our interest, despite theoretically allow precise location in space by a simple triangulation. This is due to the difficulty practice of identifying in the two images corresponding points (associated), that is, two 2-D points that really correspond to the same 3-D point of the scene.

It is insisted that the critical stage of a system The collection of delicate fruits, such as strawberry, is the precise three-dimensional determination of the point to be guided the terminal cutting element of the manipulator arm, so that it can operate with maximum cleaning only at the designated point no need to touch the fruits and at the maximum possible speed. This implies the precise detection of a given point of the peduncle of each of the ripe fruits in the row collected. In the Patent application P200501586 / 5 proposes a method based on a optical line laser for the precise designation and location of a point of an isolated peduncle, once located the fruit. In This invention, the whole problem is solved by combining the use of spot lasers to determine the position of the fruit in a cluster with that of a line laser, which is pointed in successive elevations to identify and position the desired peduncle.

The terminal element necessary for the collection is one of the prehensile-cutter type, constituted by a clamp-shaped element that grabs the peduncle at the designated point and another in the form of scissors that cut above the preset distance (as mentioned, by example, in JP2004180554). It is noted that, for this type of delicate collection, other mechanisms whose operation involves an aggression to the fruit in any of the phases (such as suction or traction).

Finally, harvesting in a greenhouse in which plants are grown in parallel rows arranged in high requires installing the vision and collection elements on a vehicle controlled by the vision system itself, capable of move longitudinally throughout the crop, so that the vision system successively access the scenes of harvest. The system could move with traction independent on rails arranged on the ground or be mounted on a sliding platform, which in turn allows access to set of rows of the greenhouse.

Detailed description

The core of the invention is based on a system of vision that allows precise guidance of a robotic arm to the apprehension, cutting and subsequent transfer and delicate deposit of each small ripe fruit in the designated position of a tray or container. The problem that allows solving is to designate with precision the exact 3-D point to which to direct the terminal mechanism prehensile-cutter that will grab the peduncle of a ripe fruit and cut it over to the preset distance The two practical difficulties it solves This invention is: first, the detection and location of the fruit to be collected, when it is presented in a spatial grouping with other fruits (cluster), and therefore is hardly distinguishable by traditional methods based only on color or shape; and, second, that of determining an exact point of your peduncle, distinguishing it from the peduncles of other fruits next not to be cut. The control and processing system It can operate on a commercial PC and allows you to manage your arm robotic taking advantage of its maximum movement capacity. Three advantages over any other solution devised until the Date for this type of accurate collection are:

1) It is capable of accurately designating the order 1 mm the position of the cut point of the peduncle of a fruit collectible, even and especially in the case of fruits that ripen in clusters (for example, strawberries), distinguishing it from the peduncles of other fruits, ripe or not, in the same cluster. Therefore, the system has a collection efficiency optimal, to the extent that it is able to collect most of the fruits of the rows (and not only those that grow isolated).

2) On the other hand, this precision allows use a scissor cutting device; the fruit, for Therefore, it can be grasped and cut by the peduncle, and transported without brush it to its destination point in the collection tray, which ensures a unique quality of the fruit collected because it is not touched or subjected to any aggression in the process.

3) Finally, the operating speed of the vision and processing system is such that the robotic arm does not have no dead timeout, which means maximum speed collection (which is not possible with other systems of view).

The basic vision system (Fig. 1) is composed of: two color CCD cameras (1) arranged preferably with parallel axes pointing to the region in the that are the fruits (5), a matrix (2) of optical laser diodes punctual (spot type) of low cost (for example, pencils or commercial laser pointers at a wavelength close to 650 nm.) mounted on an adjustable platform (3) of the type pan-tilt (that is, orientable in azimuth and elevation), which will be used to orient the matrix to the angle wanted; and an additional laser diode (4), also on the platform, which projects a flat optical beam in the instant and place designated. This commercial diode is nothing more than a spot diode, like the ones that make up the matrix, with a lens that spreads horizontally the light to project a fan-shaped beam in instead of a point or spot.

The invention, as shown in Fig. 2, together with the robotic manipulator (6) equipped with an element prehensile-cutter terminal (7), is installed on a vehicle (8), which carries a trailer with trays (9) attached to the that the collected fruits will be deposited. The vehicle, as is shown in Fig. 3, it slides on two rails (10) in the ground or mounted on a platform (11) arranged on it direction that the rows of plants (5a) and (5b) of which hang the fruits that you want to collect.

With the platform fixed in a given centered position between two rows, the vehicle is moving, steps on the rails, from one side of the greenhouse to the other, picking the fruits of one of the rows on your way out (5th) and on its way back those of the symmetrical row (5b), or alternately in the two rows only on your way out.

Auxiliary feeding systems, traction and control of devices and the computer processing and control are housed inside the vehicle, as well as the panel of parameter selection from which the parameters are preferred related to the degree of maturity required for the cut, the classification gauges, etc., according to the type and variety fruit, preferences according to the time, etc.

In one embodiment, the system carries four cameras arranged in two stereoscopic couples, independently using the left partner in the vision of the corresponding left row and the right partner in the symmetrical row on the right. In another embodiment, a only camera couple performs the vision of both rows, turning 180º in solidarity at the time required, commanded by The control system.

The operations that are performed in each position of collection (static vehicle) are those described to continuation:

First, the pair of cameras takes paths scene images; these are digitized and stored in real time. Then, the images are converted from the RGB format, in which the cameras are captured, to the HSV format, which is appropriate for efficient segmentation of the scene.

The segmentation operation consists of identify and isolate in the images the pixels that correspond to the characteristics of the fruit, distinguishing them from those others that correspond to the rest (bottom, leaves, stems, bags of plantation, etc.) In general, it is appropriate to perform a pre-segmentation based on color and hue or saturation, working on the coordinates H and S. In Each pixel determines whether the values of the H and S coordinates correspond to those with the ripe fruit (for example, determined red in the case of the strawberry), checking if found in the margins of specified values. The values of these margins, which can be entered and varied through the control panel, depend on the variety of fruit and grade of maturity specified for collection. This first processed removes pixels whose color and saturation do not correspond to those of the ripe fruit.

Once this is done pre-segmentation, whose result is a binary classification of the scene pixels in pixels of the fruit color and pixels of another color, a post-processing, consisting of eliminating pixels that appear isolated (noise), to stay only with those pixels grouped in spots or "blobs", that is, groups of connected pixels (neighbors) to each other, which correspond to the fruit or to a cluster of fruits in cluster.

The next step is to calculate the coordinates 3-D of certain singular points of the scene, with the ultimate goal of determining if the "blobs" They correspond to an isolated fruit or a group. To do this, first you mark singular points in each of the "blobs" 2-D of the two segmented stereoscopic images, in order to carry out the correspondence process ( identify, in each of the two images, the points 2-D that correspond to the same and single point space scene 3-D). In the vision system of this invention, as shown in Fig. 4, for each "blob" (12) of each of the two images, the position is calculated 2-D of its centroid (13) (geometric center in 2-D) and, drawing virtually two lines horizontal and vertical that pass through the centroid coordinates, the four extreme points of the "blob" are also marked: upper (14), lower (15), right (16) and left (17), which they will be used to determine the size of the "blob" and if it corresponds to an isolated fruit (and, in this case, determine its caliber) or to a cluster.

Then the correspondence is made between the 2-D centroids of all the "blobs" of the two images, that is, the pairs of centroids corresponding to the same spatial point (approximate). To make the correspondence of the centroids you can Use any known method. In our invention, to the point centroid of one image is matched in the other one that is at the same height (coordinate Y) and whose disparity in X (difference of its X coordinates) is compatible with the geometry of the cameras for the average viewing distance. Established the correspondence between the centroids, and therefore between "blobs", the actual coordinates in space are calculated 3-D of the apparent center of the "blob" (centroid). They are also calculated, in the 3-D space, the points corresponding to the extreme points of the "blob" before marked (upper, lower, etc.). Reconstruction 3-D of a point from its projections Stereoscopic is a trivial problem of known solution, which in the real case requires only knowing the matrix of transformation corresponding to the stereo system cameras, which is calculated in a previous calibration process.

Obviously, 3-D positions calculated must refer to a fixed coordinate origin, which it can be taken anywhere in the robotic system or the vehicle. In a prototype of this invention it has been taken as coordinate origin the central midpoint between the cameras, in the plane in which they are fixed mechanically (on the vehicle). This point will be set in the calibration process and will also be the origin of coordinates to which the arm movement will refer robotic and pan-tilt platform.

Once the positions have been determined 3-D of the characteristic points of each "blob" of the scene (centroid and extremes), each one of the "blobs" sequentially from left to right (or from right to left), as described in continuation.

First it is determined if the "blob" corresponds to a single isolated fruit or several grouped. For this compares the distances in 3-D between the upper point and lower point and between the right point and the left point of the "blob", respectively with the values of diametral reference of the type of fruit that is collected (which previously they have been introduced in the control panel and that they depend on the type of variety of the fruit).

If the comparison is consistent with the hypothesis that the "blob" corresponds to a single fruit isolated, the measures taken are used directly to estimate the caliber of the fruit for later classification, and it it proceeds to determine if its degree of maturity is the one desired for its harvest; for this the percentage of pixels for the that the coordinates H and S take values between the preset ones for the mature fruit.

If the percentage is lower than specified (selectable in the control panel), the fruit is classified as does not mature and goes on to analyze another "blob".

If the percentage is higher than specified, the fruit is classified as ripe and has to be ordered to direct the arm your pre-cutter device to a point of Fruit peduncle. The clamp will grab the peduncle, the scissor device will cut it over and transfer the fruit to the designated tray where you will deposit it in the receptacle that correspond, according to your caliber.

For this, in general, it is essential detect with precision the position of the peduncle. This is true even when the ripe fruit is isolated because it can be in the vicinity peduncles of other fruits (immature and, therefore, not detected) that should not be cut. The detection Precise peduncle is described below.

If it is determined that the "blob" does not corresponds to a single isolated ripe fruit, it is treated as a grouping of several in cluster. In that case, the procedure to individualize the fruits one by one in the cluster successively, starting with the one that is closest to the vision system

To do this, commanding the platform pan-tilt, the laser diode array is oriented towards  the grouping of fruits, using as reference guidance of the matrix axis the 3-D coordinates of the centroid of the "blob" calculated previously. The control logic of turning on / off lasers causes them to generate a grid of laser light points (marks or spots) on the "blob" of fruits, as shown in Fig. 5, for a short time, during which the scene is captured by the stereoscopic system of cameras. The next step is to accurately determine the 3-D position of each of the light points (18) of the reticle in the group of fruits (19). The light marks have a typical collection distance diameter between millimeter and millimeter and a half. In each of the two images two-dimensional taken by the cameras, the marks are isolated, using for example a segmentation algorithm similar to that described for pre-segmentation of "blobs". In that In this case, the H and S coordinates were compared with the desired thresholds. In this case, the pixels that correspond to the brand in each projection can be detected by comparing its V values with a minimum threshold, taking advantage of the fact that the brand has a luminosity considerably higher than the rest of the scene. Another option, even simpler, is to perform a subtraction operation in the V coordinate of the image illuminated by lasers and the image without to illuminate. The result is an image in which only the Lasers marks, directly segmented.

Segmented brands, their 2-D centroids in both images by the procedure described above, correspondence is made between them in the two images by the procedure also described and the 3-D reconstruction of the mark on the group of fruits. There is a point map three-dimensional on the surface of the grouping of fruits.

The more laser marks in a group, the 3-D representation of it will be better. Without However, the greater the number of brands, the more difficult the process of correspondence between the two images. To facilitate the process of correspondence between the two images without giving up a good 3-D representation of the grouping, can be use different color lasers (wavelength) interspersed: in this way, using, in addition to the V coordinate, the coordinate of color H, only the correspondence of those brands that have the same color.

About the map of three-dimensional points obtained the following processing is performed to estimate the position of each of the fruits to be collected: first, that more is taken next to the vision system. Next, determine which of 3-D points adjacent to that in the grid correspond to the same fruit, without calculating the distances three-dimensional between them and discarding those whose distance is greater than a number related to the size of the fruit.

Determined the 3-D points of the grid that belong to the same fruit, is located, with precision of the order of cm., the position of the fruit in the cluster. In concrete, the 3-D position can be determined approximate of the top point of the fruit, averaging the coordinate x (lateral) of all points belonging to that fruit and taking the y (height) and z (depth) coordinates of the point of the fruit that has a coordinate and (height) greater. This point is will correspond approximately to the place where the peduncle is born of the fruit Then proceed to locate your peduncle with millimeter order accuracy.

To accurately detect the peduncle, this invention extends the method described in P200501586 / 5 to Detection of isolated peduncles. To do this, again controlling the pan-tilt platform on which it is also located the line laser (4), this one is directed to project the line laser (20) approximately 1 cm. above the top of the fruit, using the 3-D position as a reference from the top point of the fruit calculated as described above, so that the line intercepts a point (21) of the peduncle (22), as shown in Fig. 6. Once oriented along the desired axis, the Line diode control logic feeds it, it projects the beam (23) and a spot (21) isolated explicitly appears at the point Illuminated peduncle. If the peduncle were isolated (no there would be peduncles next) the scene would be acquired with the cameras, this unique spot would be segmented, would correspond in the two images and its 3-D position would be rebuilt, thus having the almost exact coordinates of the point of the peduncle to which the mechanism is to be directed prehensile cutter.

In practice, the target peduncle is or it may be close to others of other fruits (of the cluster or of other immature not detected) and, therefore, laser lighting Line generates several brands (one per peduncle). To determine what is the target peduncle the illumination is repeated in successive elevation angles, as illustrated in Fig. 7, so that generates a trace of marks (24) for each peduncle present. The peduncle to be cut is the one for which a  simple extrapolation of the trace marks pass closer to the point superior of the fruit to be collected, whose 3-D position It had been previously calculated.

To the desired point of the peduncle, precisely determined by the previous procedure, the clamp is directed, which grab the peduncle gently and divide the fruit fractionally from the cluster to isolate it from the rest of the fruits. There is nothing left treat it as an isolated fruit as described above to isolated fruits: its degree of maturity and its size are determined and, in case it is necessary to cut it, it is cut and deposited in the corresponding receiving tray. Otherwise, it is released smoothly and the entire operation is repeated in another region of the "blob."

After each cut of fruit, the operation on the "blob" from the beginning, identifying, locating, classifying and cutting, where appropriate, the other fruits of the cluster, possibly occluded in the images taken previously. The process continues until there are no fruits left in the cluster or until the remaining ones are not collected because they have not reached the desired maturity point.

Once collected all the fruits in all the "blobs" of the scene, the vehicle moves longitudinally on the platform the distance necessary to that the cameras capture an adjacent scene on the row in harvest.

How to perform the lighting pattern of the lasers of the matrix admits several variants, all of them viable. By way of example, this part of the invention can be perform:

- using a single laser pen that, installed on the adjustable platform, it is mechanically deflected successively, and lights up synchronously with the system vision-capture. In each position the corresponding mark and the camera couple takes the projections of the scene, allowing a unique correspondence of the points of each image

Despite the simplification in the Correspondence of the spots, this mode of operation is not the recommended, because it unnecessarily slows the process of crosslinking by requiring mechanical deflection, assuming I change a slight economic advantage.

- using a matrix of lasers that light sequentially at the same speed 0 as the pickup of the cameras that take the successive images of the spots projected on the cluster. The correspondence between the points of the two projections is also direct because in each pair of images only one mark appears at a time.

- using, as described, a matrix of lasers with the same arrangement, but illuminating all of them simultaneously and making a multipoint correspondence. In In this case, if you wish to provide the correspondence algorithms and quickly identify each pair, lasers can alternate from color (red, green, blue, etc.), or be the same color but of different power, so that the different brightness of each spot allow discrimination by intensity.

Finally, the line laser can be installed in the hand of the robotic arm instead of integrated into the pan-tilt platform. In that case the aim is control by properly orientating the arm. The ignition can be performed at the time you want (for example, you can guide first arm towards the fruits and turn on the laser when the hand be next).

Brief description of the drawings

Fig. 1 Main elements of the system view. The two CCD cameras (1) point to the area where the fruits (5) to capture the scene and the matrix of point lasers (2) and the line laser (4), mounted on the adjustable platform (3), they will point to the area indicated in each case.

Fig. 2 Schematic view of the vehicle (8) on which will be placed the vision system (composed of cameras (1), the adjustable platform (3), the matrix of point lasers (2) and the line laser (4)) and the robotic arm (6) equipped with a prehensile-cutter terminal (7), which will handle collect the fruits indicated by the vision system. The vehicle it has attached a trailer with trays (9) where the harvested fruits

Fig. 3 Schematic perspective view of the movement of the vehicle (8) on the rails (10) of the platform (11) in relation to the disposition of the fruits to be collected (5th) and (5b).

Fig. 4 Example image obtained after pre-segmentation and post-processing in which shows several "blobs" (12) of different sizes. Be indicate the centroid points (13), upper (14), lower (15), right (16) and left (17) of each of them, from the which is estimated its size.

Fig. 5 Example of the grid of points (18) that projects the matrix of lasers on a group (19) of fruits in cluster.

Fig. 6 Schematic perspective view of the Basis of operation of the line laser. Fan light (23) emitted by the line laser (4) that is intercepted by the Peduncle (22) generates an isolated point (21) on it.

Fig. 7 Profile (a) and elevation (b) of the lighting in successive elevation angles that the line laser performs (4). The trace of points (24) produced in each peduncle this procedure.

Fig. 8 Diagram of the box (25) in which place the laser array (2) and the line laser (4) in a mode of realization.

Fig. 9 Schematic of the robotic arm (6) in which the grip terminal accessories (26) and cut are shown (27).

Fig. 10 Example of possible arrangement in the vehicle interior (8). The controller unit is located there of the arm (28), the feeders of the cameras (29), the platform controllers (30), the control PC (31) with the Parameter selection panel, consisting of the keyboard itself (32) of the computer and a conventional flip-up flat TFT screen (33), and wiring (34).

Fig. 11 Interconnection and exchange scheme of signals between the different blocks of the system.

Exhibition of an embodiment

All the necessary elements to perform this invention are available, commercially ready to be integrated; the algorithms necessary to perform the operations of vision can be encoded following detailed instructions in the previous section. The algorithms to control robotic arm movements are available in the literature open as reverse kinematics methods to control robots from various degrees of freedom and it is also common to be provided by the own manufacturer.

As an example of a concrete embodiment of the invention will take the case of compound vision system only by two cameras arranged to perform lateral vision with with respect to the line of movement of the vehicle and mounted on a turntable with 180º capacity to turn on itself that the cameras can realize the lateral vision in the line opposite. In this embodiment, the collection is done first in a crop line from one side of the greenhouse (beginning) to other (final). Then, turning the cameras, on the way back (from end to beginning), collection is done on the line opposite.

Lasers (punctual and line) are They can choose from the abundant commercially available. A Recommended option is to select them at a wavelength close to 650 nm, due to its low cost. Lasers, encapsulated with its lens in a cylindrical device, are arranged in the form of matrix as indicated by way of illustration in Fig. 8, in which they appear arranged in a geometry of 3 rows and 7 columns (21 spot lasers). It is noted that the laser configuration should be chosen based on the characteristics of the crop and in special depending on the maximum size of the fruit clusters. The matrix of lasers (2) is arranged in a box (25), in which It can also accommodate the line laser (4). Lasers are controlled  individually with the logic on / off, which is no more that a logic that feeds or not to each of the lasers, according to indicate the central control of the PC.

The box with the matrix of lasers is fixed on a pan-tilt platform controlled by RS232 interface from the control PC synchronously with the rest of the vision system.

The vehicle, with the attached trailer that transports the tray system, is arranged on the rails of the platform. The vehicle travels longitudinally, gathering the fruits on the platform, as indicated more above.

The movement is performed by the action of a servo motor acting conventionally on the system vehicle traction and whose displacement at each step is determined by the field of vision of the cameras chosen, so that in the next position the vision of the cameras is adjacent to the previous one. The movement order is given by the control PC, to through the servomotor controller, at the moment when The collection of each scene ends. At the end of the hall collection, there is a simple switch that activates when reached by one of the wheels of the vehicle, indicating that it has reached the end of the hall and reversing the sense of vehicle movement and rotating the camera platform 180º. The collection operation is then performed on the line of twin crop in the opposite direction until the vehicle It reaches the beginning of the greenhouse.

It is clear that the philosophy of movement between the rows can be executed in several ways depending on the greenhouse geometry

In the upper part of the vehicle the turntable with parallel cameras, and with the angle of elevation necessary to capture the collection scenes. The pan-tilt platform on which the matrix is fixed of lasers and the line laser is arranged so that it is capable of be directed to one side or the other.

Since the platforms commercial pan-tilt usually have a capacity to limited azimuth turn (for example, +160 °), must be provided so that the blind angular sector, which cannot be addressed Lasers, does not match the collection areas.

In the same way, the arm or device robotics should be arranged so that the region accessible by the end (hand with terminal elements) include the areas of collection on both sides of the vehicle and the tray area Fruit receiver in the tray carrier vehicle.

The arm (6), as shown in Fig. 9, bears in its end the grip accessories (26) and 15 cut (27), which They control independently. The prehensile device grabs first the peduncle of the fruit at the point indicated by the system of vision, approximately 10-15 mm from the fruit. He cutting device immediately cuts the peduncle to 10 mm. by over. The fruit is transported by the prehensile device of the arm and deposited in the tray or receiving basket.

Inside the vehicle they have the mode that best suits your geometry (see a possible arrangement in Fig. 10): the arm control unit (28), the feeders of the cameras (29), the platform controllers (30), the Control PC (31) with the parameter selection panel, which in this embodiment is chosen to be the keyboard (32) of the computer and a conventional flip-up flat TFT screen (33), and the wiring (34).

Interconnections and signal exchange between the blocks of the embodiment are performed as indicated by continued (see Fig. 11):

Each of the cameras is connected, through a coaxial cable RG-58-U to the corresponding commercial image capture card (type frame-grabber), in charge of digitizing images and transfer them to PC memory through the PCI bus, commanded by the main program.

The platform responsible for orienting lasers it's a pan-tilt (azimuth-lift) commercial with precision better than a tenth grade, ability to turn in elevation of between -45 and + 30º and in azimuth of + -160º, which it allows to orientate it to both sides of vision. The control of movement is done from the PC through RS232 interface, communicating the coordinates of the desired movement to the controller of the unit.

The manipulator can be an articulated arm vertical five degrees of freedom, with repeatability of movements better than + -0.02 mm. Your movements are determined by the controlling unit, to access the coordinates indicated by the PC through TCP / IP ethernet or through the serial port. He arm is mounted with a base of ISO standard terminal elements 9409, on which the two terminal elements are fixed, which are they can act electrically or pneumatically through their own arm, by the controller that executes the orders given from the control program on the PC. The terminal elements are: a Caliper with parallel fingers and a commercial cutting device compatible with ISO 9409.

Lasers are controlled from the program main through the parallel port or a PCI card signal generator (as many as lasers), which is responsible for feed lasers individually at the right time.

Industrial application

The invention has industrial application immediate used as a small fruit collection system grown in rows. It is especially suitable for collection  automated in fruit greenhouses grown in structures regular, in which a vision system guides the system robotic, which performs the tasks of apprehension, cutting and deposit of the fruit in the desired place. The invention allows to designate effectively 3-D position requires the point at which guide the terminal mechanism, which will grab the peduncle of a fruit mature and cut it over at the preset distance, distinguishing it from the peduncles of other nearby fruits that are not They must cut.

Claims (4)

1. Stereoscopic artificial vision system for precise guidance of robotic manipulator to detect, locate and collect small fruits, which can grow grouped in clusters, with the ability to isolate and determine the position of each fruit, and a point of its peduncle towards the The robotic manipulator will be guided, characterized by:

?

differentiate and locate individual fruits in clusters or groups from the three-dimensional reconstruction of the spots or brands projected in the cluster by a set of arranged optical lasers in matrix, and

?

accurately detect peduncles from the marks generated on them by the fan beams projected by a line oriented optical laser in successive elevation angles.

2. System according to claim 1, characterized in that the laser projections are achieved by deflecting a laser pointer, according to the predetermined pattern, to generate the spot matrix. 3. System according to claim 1, characterized in that the spots are generated with lasers of different wavelengths or with lasers of different powers.
Inc. 4. System according to claims 1, 2 or 3, for use in the collection of strawberries or other delicate delicate fruits grown in rows high, as in hydroponic crops in greenhouses, equipped with a robotic device with grip and cut tweezers, installed on a sliding vehicle, with container attached with trays to deposit the fruit, which performs, in each aisle, the collection in both directions of longitudinal movement, using either two pairs of cameras or only one pair with the ability to rotate 180º azimuthalm -
tea.