WO2025219011A1 - System for assisting with the cutting of plants - Google Patents

Abstract

The invention relates to a system for assisting with the cutting of plants (4), the system comprising at least one computer device (6), which is connected to at least one image capturing device (3), an image evaluation device (5) and at least one display device (7) comprising at least one light source (8), wherein at least one image of the plant (4) to be cut is generated by means of the image capturing device (3) and is analysed and segmented by the image evaluation device (5) or is examined by means of artificial intelligence in order either to project the cutting positions (14) onto the plant positions (4) by means of a light beam (13) of the light source (8) of the display device (7), or to cut the plant (4) directly at the cutting positions (14).

Description

System to assist in pruning plants

Description

The invention relates to a system for assisting in the cutting of plants, comprising a computer device connected to at least one image capture device, an image evaluation device and a display device.

For the purposes of the invention, plants are understood to mean, in particular, woody plants such as grapevines, fruit trees and fruit bushes, Rosaceae, and generally trees, bushes, shrubs, and ornamental shrubs in agriculture and forestry, as well as in parks and gardens. These are often perennial plants whose axils become woody and remain permanently intact, so that their above-ground shoot system increases in size over the years. Plants must be pruned to positively support their development, and proper pruning of plants requires expert knowledge.

Various systems have been developed to enable or automate the professional and proper pruning of plants by non-experts.

EP 3 617 816 A2, for example, discloses a computer-implemented method in which, by forming various actual and target models of a plant, instructions for cutting the plant are generated, which are displayed to the user on a portable computer device.

Furthermore, CN 102696411 B discloses a method for analyzing and processing images of grapevines. The image information captured by a binocular camera is transmitted to a computer, which performs image preprocessing, image enhancement, image segmentation, image recognition, and performs other processing steps on the image information. Based on morphological differences between the branches and trunks of the vines, the branches are identified, the pruning point is analyzed and determined, and the three-dimensional coordinate information of the pruning point is extracted. The vines are then pruned using a corresponding pruning device installed on a self-maneuvering and navigating vehicle.

CN 104871841 A describes a fruit tree pruning aid with a frame. A multidirectional video device, a remote transmission module, an input panel, and the like are arranged on the frame. A plurality of video cameras are arranged in a ring on the multidirectional video device. When using the prompt device for pruning fruit trees, the multidirectional video device captures the multidirectional, multidimensional video information of the fruit trees, which is transmitted to a remote management center via the remote transmission module. The remote management center's computer calculates an optical pruning mode and transmits the optical pruning mode back so that the pruning mode can be displayed on the prompt panel. The remote management center collects the data for pruning various fruit trees with respect to multiple shapes and purposes for future reference. Lighting components are attached to the cameras of the multidirectional video device to provide illumination as needed; the prompt panel may be an integrated tablet computer that transmits data over a network. With the fruit tree pruning aid, fruit trees can be pruned effectively using the provided pruning tips.

Finally, US 2016/0050852 A1 discloses a system for assisting in the cutting of plants, comprising at least one computer device which is connected to at least one image capturing device, an image evaluation device and at least one light source comprising at least one light source. Display device is connected, wherein the image capture device generates at least one image of the plant to be cut, which image is analyzed and segmented at the image evaluation device in order to cut the plant directly at the cutting positions by means of a light beam from the light source of the display device.

The invention is based on the object of creating a system for assisting in the cutting of plants of the type mentioned at the outset, which provides a user with assistance in cutting plants with little effort.

According to the invention, the object is achieved by the features of the independent claim.

The subclaims represent advantageous embodiments of the invention.

A system for assisting in the pruning of plants comprises at least one computer device connected to at least one image capture device, an image evaluation device, and at least one display device comprising at least one light source. The image capture device generates at least one image of the plant to be pruned, which image is analyzed and segmented by the image evaluation device or examined using artificial intelligence in order to project cutting positions onto the plant using a light beam from the light source of the display device and/or to directly cut the plant at the cutting positions. In particular, for example, a 3D model of the plant to be processed can be created.

This invention relates to horticulture, fruit-growing, and viticulture and describes a system that simplifies the professional and proper pruning of plants with the support of cameras, sensors, computers, and light, so that even laypeople can perform such work almost error-free. The system is in Capable of collecting and processing information about the plants to be cut, and using a light beam to mark the areas to be cut on the plants or, if a suitable light source is used, to cut them directly. The light beam is usually a laser beam.

For example, at least one actuator or control element connected to the computer device can align the at least one light source such that the respective light beam marks a cutting position. At least one marking laser, also known as a pointer, is used to mark the points to be cut on the woody plant. At the calculated cutting position indicated by the light beam, the plant is cut manually, for example, using scissors or a saw.

The computing unit controls the display device via interfaces to align the emitted light beam in order to apply the cutting positions as corresponding light projections onto the plant.

If necessary, the display device has at least one alignment device with corresponding adjustment elements and/or at least one optical device to precisely align the light beam to the calculated cutting positions. As light sources, the display device comprises, in particular, at least one marking laser with a relatively low energy density and/or a cutting laser suitable for cutting the plant. The optical device can comprise, for example, mirrors or light guides.

Alternatively or in addition to the at least one marking laser, at least one cutting laser can be installed in the display device, which cuts the plant directly at the respectively calculated cutting point, which can optionally also be additionally displayed by the associated marking laser. The cutting laser can preferably be a A controllable laser, particularly a CO2 laser or another state-of-the-art laser, can be used. For example, a UV diode laser can also be used as a cutting laser, which is robust, inexpensive, and compact.

The expert knowledge within the scope of this invention is implemented in the form of computer programs that run on the devices, i.e., at least one computer device. The computer programs include both algorithmic methods and artificial intelligence, e.g., in the form of neural networks.

A computer device within the meaning of this invention includes both mobile devices such as smartphones, portable microcomputers, embedded systems, single-board computers, and microcontrollers, as well as stationary computers, servers, and PCs on which the computer programs run. The at least one computer device involved in the system is also referred to below as the computing unit.

This invention differs from the relevant prior art regarding plant pruning or support for pruning vines, particularly in the way the results are displayed or applied. Previously, pruning suggestions were displayed using AR or VR headsets or on screens. In contrast, the present invention applies the cutting positions directly to the plant to be pruned using at least one light beam. In a particular embodiment of the invention, the present system cuts contactlessly using laser light.

The image capture device comprises at least one camera or at least one lidar sensor for capturing images of the plant to be cut. The camera can operate in the visible and/or invisible wavelength spectrum of light to achieve the best possible image of the plant in its current state. If multiple cameras are used on the image capture device, the quality of the camera images can be improved by combining them. and/or three-dimensional images are generated. The camera images can be recorded as single images, video, or a data stream.

The above statements regarding the camera apply to the LIDAR sensor, which is designed, for example, as a time-of-flight camera, i.e., one based on a time-of-flight measurement. With LIDAR technology, the plant to be cut can be scanned using laser light, and very precise three-dimensional measurements can be determined. The LIDAR sensor is used in particular with the sensors described below, i.e., preferably in conjunction with a satellite navigation system and/or a distance measurement system. An INS (inertial navigation system) can also be used to measure the roll and pitch angles as well as the orientation of the LIDAR system. In combination with the position information from the satellite navigation system and the inertial navigation system, the distance measurements are used to determine a true three-dimensional image of the reflecting target, i.e., the plant in object space. The point data is post-processed after LIDAR data acquisition and integrated into a highly accurate georeferenced XYZ coordinate system by analyzing laser period, laser scanning angle, GPS position, and INS information.

Preferably, distance determination sensors are used to improve the quality of the calculation results. Such distance determination sensors can also be arranged, in particular, alternatively or in addition to LIDAR sensors, whose sensor data are processed on the computer device as explained above.

Furthermore, sensors of a satellite navigation system are preferably provided for determining the position in space, which are designed, for example, as GPS sensors. Furthermore, at least one gyro sensor is preferably provided to determine the orientation and, if necessary, movement in space.

The aforementioned components, in particular the image capture device and the sensors, are used for data acquisition and are deployed in a mobile manner near the plant being treated. They are referred to below as data acquisition components.

The sensors can also be assigned to the display device for position data acquisition in order to calibrate it or to more precisely align the light beams of the marking laser and/or the cutting laser.

The data from the data acquisition components is transmitted via interfaces to the computer device for further processing. These interfaces can be internal (e.g., smartphone) or external (e.g., Wi-Fi, Bluetooth), wired or wireless, without departing from the scope of the invention.

The processing unit or computer system processes the data received from the data acquisition components into positions for cutting on the plant. For this purpose, the plant images are segmented by the image analysis system to identify the components relevant for calculating the cutting positions. Segmentation can be performed using conventional image processing methods or artificial intelligence.

In addition to segmentation, 3D modeling is performed, and the intersection positions are calculated from the combination of segmentation and 3D modeling. Likewise, as an alternative to segmentation, the intersection positions can be calculated directly from the camera and sensor data using artificial intelligence, for example, deep learning or another suitable method, without segmentation and 3D modeling. A three-dimensional model (3D model) of the plant is calculated from the plant images and/or segmentation, possibly supplemented by data from the sensors coupled to the computer system. A three-dimensional graph (3D graph) is derived from this 3D model for further calculations.

The cutting rules are stored electronically in at least one memory device of the processing unit and can be implemented algorithmically, e.g., as a decision tree, a neural network, or an analog computer. The cutting positions are then calculated from the cutting rules and the 3D graph. Alternatively, the cutting positions are calculated using artificial intelligence (e.g., a neural network), which was created using deep learning and may require no 3D modeling.

The computing unit controls the display device via interfaces to align the emitted light beam in order to apply the cutting positions as corresponding light projections onto the plant.

If necessary, the display device has at least one alignment device with corresponding adjustment elements and/or at least one optical device to precisely align the light beam to the calculated cutting positions. As light sources, the display device comprises, in particular, at least one marking laser with a relatively low energy density and/or a cutting laser suitable for cutting the plant. The optical device can comprise, for example, mirrors or light guides.

The data from the aforementioned sub-steps are processed on a mobile computer device, which can form a structural unit with the data acquisition components, in particular the image acquisition device and preferably the sensors. This unit can in particular be a smartphone or It can be an embedded system or a mobile PC that controls the pointer via interfaces and is referred to below as a mobile component.

It is also possible to implement it as a distributed system, in which the mobile component is expanded with additional computing units or computer devices in order to perform functions for which the mobile component is not suitable on its own due to its inherent performance or computing power and storage capacity. These can be particularly computationally intensive processes that are executed on one or more stationary computing units, e.g., servers with many hardware resources (CPU, GPU, RAM, read-only storage). All possible network interfaces and protocols suitable for enabling data exchange between computing units or computer devices can be used as interfaces for the distributed system. Depending on its size, weight, and energy requirements, the mobile component can be designed as a portable device or mounted on a ground-based mobile platform, for example with a chassis, in particular with or without its own propulsion, or a drone, hereinafter referred to individually or collectively as a mobile device.

The mobile component can, if necessary, influence the motion control of the mobile via interfaces, thus creating an autonomously operating mobile. This allows plants to be trimmed automatically without human intervention. Multiple mobiles can be linked via interfaces to form a swarm of mobiles if necessary. This allows many plants to be trimmed simultaneously in a short period of time.

If the mobile component is worn by a user, preferably attached to their body or work clothing, the user monitors the cut or performs it with a suitable tool, e.g., plant shears. If the plant shears are electrically or pneumatically powered with a drive control, these plant shears can be connected to the mobile unit via interfaces. The mobile unit can The control particularly takes over the opening and closing of the plant shears if the drive can be controlled accordingly.

The light source is calibrated using the cameras, sensors, and the processing unit. The display device can, of course, have its own sensors or cameras, particularly for calibrating the light beam or light source, and can operate independently of the image processing device, for example. The cutting positions are used to control the light source so that the light beam is directed at precisely the parts of the plant that are to be cut. Additionally, the light source can be used to project information onto the plant or into the surrounding area, e.g., if certain information is still missing for the cut calculation, or if the system user or wearer of the mobile component should assume a different position for the cut. This is particularly the case if, from a certain position, branches of the plant lie one behind the other or if the cutting positions cannot be optimally projected onto the plant. Depending on the power and energy content of the light, the light source displays the cutting positions as markings on the plant or cuts them directly.

The mobile unit can be powered by a built-in battery, an external connected battery (e.g. a so-called power bank or similar), the power supply of an electric drive of the plant shears, portable solar cells, portable generators or the power supply of the mobile.

Finally, it is proposed that a cutting laser, which, for example, has sufficient power to cut off particularly woody parts of a plant, such as grapevines, be movable and/or alignable in three dimensions on its own movable arm. This means that the cutting laser can be aligned and positioned in space accordingly in order to make a cut at a desired location, in particular controlled by a corresponding evaluation program. Thus, For example, woody, dead, or other unwanted parts of a plant are to be cut away. In particular, optical monitoring can be achieved using a marking laser, which projects an optical mark at a specific location so that the process can also be directly monitored by a user. This means that a mark is projected and then automatically cut at the desired location with the cutting laser. This can be monitored by a user.

For this purpose, the system can comprise a marking laser, a cutting laser, and two mirrors. The cutting laser, for example, a high-energy UV laser with a short wavelength, shines through a mirror that is semi-transparent for the corresponding wavelength. The light from the marking laser, for example, a lower-energy laser with red, visible light, is reflected off this semi-transparent mirror, and the two laser beams are projected together by a second mirror onto the desired cutting point. This second mirror is designed to be movable and can be rotated and/or tilted in order to direct the two light beams in the desired direction. Several second mirrors can also be provided if this should be necessary due to the shape of the plant to be treated. First, the desired cutting point can be marked with the marking laser, and then, if necessary after user approval, the cutting laser is activated to cut at the desired location. This can also be automatically checked by the system via optical detection using image processing. For example, the two lasers and mirrors can be mounted on a common, movable robot arm to mark and/or cut any part of a plant or growth. The lasers can also be mounted separately on different, movable arms.

For example, LIDAR sensors can be used to detect the position of a desired interface. A 3D model, in particular, can also be used. dell of the plant can be calculated in order to determine a cutting position or stored cutting rules such as thickness, degree of lignification, possible shading of other parts of the plant or the like can be taken into account.

It is understood that the features mentioned above and those to be explained below can be used not only in the respective combinations specified, but also in other combinations. The scope of the invention is defined solely by the claims.

The invention is explained in more detail below using exemplary embodiments with reference to the accompanying drawings.

It shows:

Fig. 1 is a schematic representation of a housing to be assigned to the body of a user with components of the system according to the invention,

Fig. 2 a first representation of the user with the housing according to Fig. 1 ,

Fig. 3 a second representation of the user with the housing according to Fig. 1 ,

Fig. 4 is a representation of the user according to Fig. 3 in relation to a plant to be cut,

Fig. 5 is a representation of a chassis designed as an over-row frame with associated components of the system according to the invention,

Fig. 6 is a first representation of components of the system according to the invention, Fig. 7 shows a first alternative representation of components of the system according to the invention,

Fig. 8 shows a second alternative representation of components of the system according to the invention,

Fig. 9 shows a third alternative representation of components of the system according to the invention,

Fig. 10 is a further illustration of the chassis designed as an over-row frame with associated components of the system and a towing vehicle,

Fig. 11 an alternative representation of the chassis with associated components of the system,

Fig. 12 shows another alternative representation of the chassis with a frame with associated components of the system,

Fig. 13 shows a fourth alternative representation of components of the system according to the invention,

Fig. 14 a flow chart of the procedure and

Fig. 15 a system with marking laser, cutting laser and two mirrors.

Fig. 1 and Fig. 6 show a relatively simple embodiment of the system with the housing 1, in which an image capture device 3 comprising at least one camera 2 is arranged for capturing images of a plant 4, which are analyzed by an image evaluation device 5 of a connected computer device 6. The image evaluation device 5 of the computer device 6 calculates cutting positions 14 from the image data, i.e., locations at which the plant should be trimmed. For this purpose, the image is segmented and 3D modeled, followed by the calculation of the cutting positions by combining the segmentation and the 3D model. Alternatively, the cutting positions can be calculated directly and without segmentation and 3D modeling using artificial intelligence, in particular deep learning or another suitable method.

The housing 1, in particular made of metal, or a housing, can be coated and/or matted, in particular on the inside and/or outside, for example with sand or the like, in order to reduce disturbing reflections of sunlight on the one hand and to minimize or prevent the escape of unwanted laser radiation on the other hand.

Overall, the housing 1 can be arranged, for example, on a particularly self-propelled and self-steering over-row frame in order, for example, to work a row of vines.

As an alternative to the camera 2, at least one lidar sensor 19 can also be used to create an image of the plant 4. Of course, the arrangement of at least one camera 2 and one lidar sensor 19 on the image capture device 3 and their coupling to the image evaluation device 5 of the computer device 6 is also possible.

Furthermore, a display device 7 is connected to the computer device 6, which generates at least one light beam 13 marking a cutting position 14 on the plant 4. The display device 7, which is also referred to as a pointer, comprises at least one controllable light source 8 and is controlled by the computer device 6 according to the calculated or determined interfaces. Of course, actuators and/or optical devices (not shown in detail) can be provided for positioning the pointer and the associated alignment of the light beam 13, which are also connected to the computer device 6 via suitable interfaces. Of course, an electronic module 18 is provided, which, for example, manages the power supply of the individual components, wherein In particular, batteries, solar panels or the like, not shown in detail, may be provided to provide electrical energy.

Furthermore, several sensors 10 connected to the computer device 6, which of course also includes a memory device 9 in addition to a processor, are arranged in the housing 1, the sensor data of which are evaluated by the computer device. The sensor data is used to determine the cutting positions 14 and to display them on the plant 4 using the alignable light beam 13 of the display device 7. The sensors 10 can be a distance sensor, a sensor of a satellite navigation system, and/or a gyro sensor. After calibration of the sensors 10, the sensor data, in addition to the image data, describe the orientation of the housing 1 in the environment and in relation to the plant 4 and ensure the precise alignment of the light beam for marking the cutting position 14 on the plant 4.

In one embodiment, the system according to Fig. 2 and Fig. 3 is equipped with the image capture device 3 comprising two cameras 2 or two lidar sensors 19 or one camera 2 and one lidar sensor 19, which enable stereo images of the plant 4. The housing 4 is arranged by means of a fastening strap 11 in a chest area or forehead area of a user 12 of the system such that the cameras 2 or the lidar sensors 19 and the display device 7 are aligned in the line of sight. Images of the plant 4 arranged in front of the user 12 are thus recorded and evaluated, and the light radiation is projected onto the cutting points of the plant so that the user 12 can immediately perceive it. The user 12 has both hands free and can cut the plant 4 at the marked cutting positions 14 using plant scissors.

Fig. 4 illustrates the display of a cutting position 14 on the plant 4, which is located in the viewing direction of the user 12 by means of the laser beam acting as a marking laser. formed light source 8 emitted light beam 13 of the display device.

If the display device 7 comprises a light source 8 designed as a cutting laser, which can be arranged in addition to or alternatively to the marking laser and, for example, has a higher energy density, a cutting laser beam 21 is emitted, which can cut the plant 4 at the determined cutting position 14, thus eliminating the need for manual processing. Naturally, the light source 8 designed as a cutting laser can be aligned by means of controlled actuators and/or the cutting laser beam 21, like the marking laser beam 22, can be deflected by means of suitable optical devices, which can also be aligned in a controlled manner, in order to reliably reach the cutting position. The transmittance of the optical device can be adapted to the cutting laser or the marking laser.

According to Fig. 5 and Fig. 10, the housing 1 can also be arranged on a chassis 16 designed as an over-the-line frame 15, which is either towed by a towing vehicle 17 or is self-propelled. The chassis 16 offers the possibility of arranging, for example, optical devices of the display device 7 or the display device 7 with corresponding actuators for aligning the light beam 13 outside the housing 1. Of course, several over-the-line frames 15 can also be arranged on a chassis 16. Alternatively, according to Fig. 12, the chassis 16 carries a simple frame 20, on which, for example, an optical device of the display device 7 is arranged to align the light beam 13.

According to Fig. 7, the image capture device 3 of the system comprises a plurality of cameras 2, which are connected to the computer device 6 and the image evaluation device 5 provided thereon. Furthermore, the display device 7 is shown in addition to the electronic module 18. According to Figs. 8 and 9, several cameras 2 and several sensors 10 are connected to a central computer device 6, which, of course, also communicates wirelessly with additional computer devices 6.n via suitable interfaces. The additional computer devices 6n can increase the computing power or are assigned to other autonomous systems, so that several plants 4 can be cut simultaneously, especially if the chassis 16 move autonomously and the light beams 13 for cutting the plants are designed as laser light.

Specifically, cameras 2 are used to capture image data of plant 4, and sensors 10 are used to determine data for orienting the system to assist in cutting plants 4. The image data of plant 4 is semantically segmented at the image evaluation device 5 using conventional image processing methods or artificial intelligence. The sensor data is used to calculate a three-dimensional reconstruction. The cutting positions and the data for projecting the light beam are calculated from the processed image data and sensor data, and the light beam is controlled accordingly.

Furthermore, the image data and sensor data are used to capture, for example, the position and movement of the system. The position data is used in particular to calculate the cutting position, and the tracking data from the system's movement is used to calculate the target position for the light beam to represent the cutting position or the cutting itself. Furthermore, the image data and sensor data are used for motion control.

According to Fig. 11, the towed or self-propelled chassis 16 carries a frame 20 on which, viewed in the direction of travel, several image capture devices 3 are arranged one above the other in order to capture image data of the entire plant 4, for which purpose the image capture device 3 is provided with cameras 2 and/or LIDAR- Sensors 19 are assigned. The image data are evaluated and processed at the image evaluation device 5 of the computer device 6. The display device 7 connected to the computer device 6 is arranged at a distance from the quasi-leading image capture devices 3 and is also connected to the computer device 6. Furthermore, various sensors 10 are assigned to the display device 7 in terms of location and data technology. These sensors 10, in addition to a camera or a LIDAR sensor, also include, for example, a satellite navigation sensor, a gyro sensor, and the like, in order to compare the position of the display device 7 with the data from the image capture devices 3 and to calibrate the display device 7 based on the sensor data. The display device 7 in the present case comprises a first light source 8, which is designed as a marking laser and emits a light beam for marking a cutting position 14 via an associated optical device, and a second light source 8 of higher energy density, which is designed as a cutting laser and sends a cutting laser beam 21 to the marked cutting position 14, wherein, of course, optical devices and/or actuators for aligning the cutting laser beam 22 can also be provided here.

According to Fig. 13, several cameras 2 and several sensors 10 are connected to a central computer device 6. Furthermore, a display device 7 with several light sources 8 is coupled to the computer device 6, wherein individual light sources 8 emit light beams 13 designed as cutting laser beams 22 for cutting the plants and other light sources 8 emit light beams 13 designed as marking laser beams 22 for marking cutting points 14.

Finally, a flow chart is shown in Fig. 14 to illustrate the interaction of the various components of the system.

In the part 100 of the system, cameras and sensors are arranged, such as, among others, an optical camera 101, a lidar sensor 102 or a TOF camera, which operates according to the "time-of-flight" principle, i.e., a time-of-flight measurement of an electromagnetic signal reflected by the plant, an IMU or position sensor 103, which can, for example, detect an inertial movement of the entire system or parts thereof, or a known GPS or position sensor 104. These sensors serve to detect the position of the system relative to a plant to be treated, such as a grapevine. For example, image data 105 is recorded, an orientation of the system 106 relative to the plant to be treated is recorded, an absolute position in space 107 is determined, and/or a relative movement 108 in space is taken into account, for example, on a self-propelled over-the-line frame.

In part 140 of the system, a calculation is performed based on the acquired data and measured values. In step 109, for example, image data 105 can be segmented, as indicated by the connecting arrow. Furthermore, in step 110, a 3D reconstruction of the plant can be created from this. From this, a desired cutting position can be calculated 111, and then, in step 112, a desired projection location for a marking laser beam is determined. This is based on 3D models 113 of plants or the respective cutting specifications or rules stored in a database.

In part 150 of the system, the position of the system is calculated, particularly relative to a plant to be processed. In step 114, a movement of the system, particularly in space, can be tracked or traced 114, and a desired target position can be determined 115 from this, for example, in order to optimally position a marking and/or cutting laser.

In part 160 of the system, various actuators are then controlled. For example, in step 116, the light beams The marking and/or cutting laser is controlled in step 117, and the entire system is moved in step 118. A 3D robot arm for the cutting laser can also be controlled accordingly. For example, one or more light sources or lasers can be controlled in step 118, a control system's optics in step 119, the system's steering in step 120, and its drive in step 121.

The connecting arrows indicate direct influences or controls.

In Fig. 15, the system comprises two mirrors 26, 27 and a marking laser 24 and a cutting laser 23. The cutting laser, for example a high-energy UV laser, emits a cutting laser beam 21, shown here in dashed lines, and the lower-energy, for example a red marking laser 24, emits a marking laser beam 22, shown here as a solid line. The marking laser beam 22 is reflected by a semi-transparent mirror 26, while the cutting laser beam 21 passes through the semi-transparent mirror 26. Behind it runs the common beam path 25 of both laser beams. At the second mirror 27, the two laser beams are reflected together and projected onto a common desired interface. At least the second mirror 27 is designed to be rotatable and/or tiltable in order to be projected onto a desired and calculated interface. The rotating/tilting can be carried out in the millisecond range. For example, the marking laser 24 can first be activated to mark the desired cutting point. This can be verified automatically, for example, by image processing and/or by a user. Subsequently, the cutting laser 23 is activated by the user and/or automatically, and the plant or growth is cut at the cutting point.

The system can be arranged entirely or at least partially on a movable 3D robot arm or distributed across several robot arms in order to be able to reach all desired locations on the plants. Reference symbol

15 overline frames

16 chassis

17 towing vehicle

18 Electronic module

19 LIDAR sensor

20 frames

21 Cutting laser beam

22 Marking laser beam

23 cutting lasers

24 marking lasers

25 common beam path

26 semi-transparent mirror

27 mirrors

Claims

Patent claims 1 . System for assisting in the cutting of plants (4), comprising at least one computer device (6) which is connected to at least one image capture device (3), an image evaluation device (5) and at least one display device (7) comprising at least one light source (8), wherein the image capture device (3) generates at least one image of the plant (4) to be cut, which image is analyzed and segmented at the image evaluation device (5) or examined by means of artificial intelligence in order to project cutting positions (14) onto the plant (4) by means of a light beam (13) of the light source (8) of the display device (7) and/or to cut the plant (4) directly at the cutting positions (14). 2. System according to claim 1, characterized in that at least one sensor (10) for distance measurement is connected to the computer device (6) for evaluating the sensor data. 3. System according to claim 1, characterized in that at least one sensor (10) of a satellite navigation system is connected to the computer device (6) for evaluating the sensor data. 4. System according to claim 1, characterized in that at least one gyro sensor is connected to the computer device (6) for evaluating the sensor data. 5. System according to claim 1, characterized in that the segmentation of the image of the plant (4) to be cut is carried out by means of image processing methods or the analysis of the image is carried out by means of artificial intelligence, wherein cutting rules are stored in a memory device (9) of the computer device (6) and implemented algorithmically or as a neural network or as an analog computer. 6. System according to one of claims 1 to 5, characterized in that the light source (8) of the display device (7) is designed as a marking laser or a cutting laser. 7. System according to one of claims 1 to 6, characterized in that the display device (7) comprises at least one optical device associated with a light source (8) and/or at least one actuator for influencing the orientation of the emitted light beam (13) or cutting laser beam (22). 8. System according to one of claims 1 to 7, characterized in that the display device (7) projects information perceptible to the user (12) about the cutting positions (14) onto the plant (4) or into the surroundings. 9. System according to one of claims 1 to 8, characterized in that components of the system are installed in a housing (1) to be assigned to the body of a user (12). 10. System according to one of claims 1 to 8, characterized in that components of the system are assigned to a chassis (16). 11 . System according to claim 10, characterized in that the chassis (16) comprises an over-row frame (15). 12. System according to one of claims 1 to 8, characterized in that hardware components of the system are assigned to an aircraft. 13. System according to one of claims 1 to 12, characterized in that the computer device (6) communicates via interfaces with further computer devices (6.n). 14. System according to one of claims 6 to 13, characterized in that a cutting laser is movable and/or alignable in three dimensions on its own movable arm. 15. System according to one of claims 6 to 14, characterized in that the cutting laser (23) and the marking laser (24) are assigned two mirrors (26, 27), of which at least one (27) is designed to be movable. 16. A swarm of systems according to any one of claims 1 to 15, wherein the systems communicate with each other via interfaces.