EA046164B1 - METHOD OF PREPARATION FOR LUGGING USING AN UNMANNED VEHICLE, UNMANNED VEHICLE AND SYSTEM USING THE SAID METHOD - Google Patents

Description

Technical field

The present invention relates to a method of preparing for logging in which information associated with objects in a forest area is obtained and each object is assigned an object identifier to prepare for harvesting trees in a forest area.

State of the art

Harvesting trees at the optimal time is a skill with a long tradition in forestry. The selection of trees for harvest is usually carried out as a multi-step process in which, as a first step, information about the forest or forest area is collected, for example from forestry maps and/or as a result of a forest surveillance operation. Forest maps and/or forest surveillance operations can be used to select forest areas for logging; a forest zone in which the forest is considered to be suitable for harvesting. Identification of individual tree objects that will be cut down in the selected forest zone is carried out as a second stage, in which decisions are made for individual tree objects within the forest zone. Selection involves determining which individual tree objects are to be harvested and performing marking operations on specific tree objects. Selecting individual tree features can also be done to identify tree features that should be excluded from logging, for example for cultural heritage or biodiversity purposes.

According to the prior art, preparation for logging of a forest area, for example including the selection of individual tree objects, is a labor-intensive process that is carried out on foot, i.e. through foresters walking around the selected forest zone. Foresters prepare for logging in an area by identifying and marking tree features, for example, by marking tree features that meet the criteria defined for trees to be harvested or the criteria for trees to be preserved, or by marking other features that should be avoided by the harvester. The determination process is highly dependent on the skill and experience of the forester. Consequently, current solutions for selecting objects, such as tree objects, within a selected area are usually labor-intensive and expensive. In addition, current solutions result in labeling that can be perceived as arbitrary and with wide variations in terms of quality and consistency.

Another tree harvesting scenario that involves marking individual objects relates to the need to reconnoiter the route before the harvester enters the selected forest area. In accordance with the scenario described above for selecting individual tree objects, foresters also identify and mark blocking objects on the terrain of the selected forest zone. Objects are manually marked to improve the harvester's navigation through the area, as well as to enable harvesting route planning, for example in terms of a preferred route or a route to avoid.

At the moment, there are no methods or systems available that can prepare for harvesting in a forested area and provide all the necessary information in terms of how and what to harvest, and in terms of biodiversity aspects and routing aspects of the area.

The essence of the invention

The purpose of the present invention is to eliminate or at least minimize the problems described above. This is achieved by a logging preparation method, an unmanned vehicle capable of logging preparation, logging systems and software in accordance with the accompanying independent claims.

A method for preparing for logging includes using an unmanned vehicle configured to move under a forest canopy in a forest area, and for each of at least one object in the forest area, the method further includes obtaining, using at least one sensor, the unmanned vehicle means, information associated with the object. The method further includes assigning an object identifier to an object based on the received sensor information using processing circuitry contained in or available to the autonomous vehicle. The method also includes associating a token with an object and the received sensor information or an object identifier assigned to the object.

Thus, preparation for logging can be carried out by an unmanned vehicle such as a drone. Sensor information corresponding to the properties of each object can be collected by the unmanned vehicle, and by assigning an object ID and associating a marker with the object and the received sensor information or the object ID assigned to the object, it is possible to make logging decisions and carry out logging operations within the forest area without the help of a forester. This results in savings in time and costs in preparation for logging, which also improves the reliability and consistency of information collected and logging decisions made, so that the extent of forest use in a forested area increases while maintaining biodiversity and preserving sensitive areas such as heritage sites.

- 1 046164

Accordingly, the method may further include obtaining a logging decision using a processing scheme based on the received sensor information or an object identifier assigned to the object, and associating the marker also with the logging decision. The harvesting decision may be made in connection with the acquisition, using at least one sensor of the unmanned vehicle, of information associated with the object, i.e. while the driverless vehicle moves under the forest canopy. Alternatively, the harvesting decision could be made later, optionally at a remote location. In some embodiments, the logging decision may be made using a processing circuit that is contained in or accessible to the autonomous vehicle, but in other embodiments, logging decisions may be made by another processing circuit or by a human operator that uses the information and assigned identifiers objects to make a decision on each object. The harvesting decision may be based on received sensor information or an object ID assigned to a single object, or received sensor information and/or object IDs assigned to multiple objects.

In this way, harvesting decisions are made automatically by the unmanned vehicle or system in accordance with the embodiments provided herein, which further contributes to the benefits of achieving time-efficient and cost-effective harvesting preparations that also improve the reliability and consistency of the information collected and decisions made about logging, with the amount of forest use in the forest area increasing while conserving biodiversity and preserving sensitive areas such as heritage sites

Accordingly, assigning an object identifier to an object using the processing circuit includes analyzing sensor information to determine at least one property of the object and selecting an object identifier from a set of pre-stored object identifier alternatives based on the at least one property of the object. Such stored object identifier alternatives could be that the object should be harvested or, alternatively, the object should be bypassed. By using pre-stored Object ID alternatives, you can have a very cost-effective and simple way to prepare for harvesting that still provides enough information to make logging decisions and execute logging operations based on the Object IDs assigned to the objects.

Alternatively, assigning an object identifier to an object using a processing circuit includes analyzing sensor information to determine a unique pattern associated with the object, generating a unique object identifier based on the determined unique pattern; and assigning an identifier to the object. The unique pattern may be a reproduction of the bark pattern of an object, where the object is a tree, or alternatively may be a reproduction of a branch pattern or graphical plant identifier, a color pattern, or any other pattern that is unique to the object. Thus, each object to which the method is applied is uniquely identified and associated with the received sensor information, so that properties of the object can be uniquely associated with the object. When harvesting in a forested area, the association of sensory information with a unique object identifier allows the harvest of each tree to be tailored to what is considered optimal for that particular tree, so that the harvest yield is increased.

Alternatively, assigning an object identifier to an object using a processing circuit involves analyzing sensor information to determine a unique position associated with the object, generating a unique object identifier based on the determined unique position, and assigning that identifier to the object. This makes it possible to very effectively associate the received sensor information with the specific location in which the object is associated, so that logging is further improved.

Accordingly, the method further includes storing information associated with the at least one object, along with an object identifier and a corresponding token, in a memory accessible to the autonomous vehicle. In this way, digital markers associated with each feature can be stored and can also form a point cloud or map of the forest area containing markers for each feature and information, and possibly associated logging decisions. Forest areas can then be harvested using digital markers alone, or digital markers in combination with physical markers in the forest, to ensure that the desired trees are harvested in the desired manner and that certain features or locations are not touched by the harvester.

Preferably, the method also includes that the harvester recognizes the marker using at least one harvester sensor, and that the harvester also receives a harvesting decision for an object associated with the recognized marker, and performs a harvesting action based on the harvesting decision associated with an object. Thus, logging can be done more efficiently by preparing an unmanned vehicle.

- 2 046164

Accordingly, the method also includes the harvester recognizing the marker using at least one sensor of the harvester, obtaining a harvesting decision for an object associated with the recognized marker, and performing a harvesting action based on the harvesting decision associated with the object. In this way, logging can be performed in a more efficient manner without requiring the human operator to recognize the marker and harvest objects in a specific manner. The harvester can be made completely autonomous, but alternatively it can have a human operator available to perform some harvesting operations while allowing the harvester to operate autonomously in other respects, or can be controlled by a human operator at a remote location, but can also be customized to recognize markers and harvest trees in the desired manner.

Accordingly, each of the at least one object is a tree or part of a tree, an area boundary, an area not suitable for harvester travel, an existing path in a forested area, an ancient remains or monument, a fallen tree of environmental heritage value, or other biological object protection. A logging decision may be a decision to harvest an item, how to harvest an item, not to harvest an item, or to avoid a location where an item is located or a specific area in which an item is located.

The method may also include generating a 2D or 3D representation at the scale of at least a portion of the forest area, or updating a 2D or 3D representation at the scale of at least a portion of the forest area by adding a marker associated with an object to the 2D or 3D representation. A two-dimensional or three-dimensional representation will hereafter be referred to as a map. Thus, by creating or updating a two-dimensional or three-dimensional representation, the map contains markers identifying features that will serve as the basis for a subsequent harvesting operation, so that each marker shows where the feature is located, giving the harvester detailed information about how to harvest in forest area.

Accordingly, a token is a physical token, and associating a token with an object includes attaching a physical token to an object or a location in the vicinity of an object, or placing a token on or near an object. In this way, the marker becomes visible in the forest area so that the harvester can harvest without the need for maps or for the information presented on the map to be combined with physical markers that are visible on site.

Proximity, which may also be referred to herein as close proximity, is established or predetermined based on what degree of precision is required in a particular logging preparation or logging process. In a non-limiting example, the (immediate) proximity of an object may be defined as greater than or equal to 0 m, or greater than or equal to 0.01 m, and less than or equal to 2 m from the object. In some additional non-limiting examples, (immediate) proximity to an object may be defined as greater than or equal to 0 m, or greater than or equal to 0.01 m, and less than or equal to 1 m from the object, less than or equal to 5 dm from the object, less than or equal to 2 dm from the object or less than or equal to 1 dm from the object. In comparison, Global Position System (GPS) technology only provides position estimates with an error of about 20 m, which is clearly not enough to reliably distinguish one object (tree, rock, etc.) from another neighboring object in a forested area. This is why prior art solutions that use GOS positioning or similar positioning techniques typically relate to identifying a group of trees, an area with certain properties, etc., rather than the positions of individual objects such as trees.

The token may be a digital token, and associating the token with an object and, optionally, a harvesting decision may include storing the token along with an identifier of the object and, optionally, the harvesting decision in memory, which can then be easily accessed by memory access before or during logging time.

The object could be a tree that needs to be cut down, and making a harvesting decision based on the sensor information received could include:

calculating, by a processing circuit, a three-dimensional shape of a tree trunk based on the received sensor information; and calculating, by the processing circuit, an optimal method of cutting the tree trunk at the at least one identified location to maximize the possible yield of high quality wood for the tree, wherein associating the marker with the object and the harvesting decision may include physically marking each of the at least one an identified location on the tree trunk or digitally marking each of the at least one identified location on the tree trunk in a digitally scaled three-dimensional representation of the tree trunk. In this way, detailed information is obtained on how to optimize wood harvesting, and using this information, timber harvesting can be carried out efficiently.

According to the invention, an unmanned vehicle is provided that can be prepared for logging. The unmanned vehicle contains a body made

- 3 046164 with the ability to move through a forested area under the forest canopy, and at least one sensor configured to receive information associated with the object, wherein the unmanned vehicle contains or is communicatively coupled with processing circuitry configured to assign an object identifier to the object based on the received sensor information, wherein the unmanned vehicle is further configured to associate the marker with the object and the received sensor information and/or an object identifier assigned to the object. In this way, a self-driving vehicle can prepare for logging by moving through a forested area so that later logging can be done efficiently in the same area.

Preferably, the processing circuitry is further configured to make a harvesting decision based on the received sensor information and/or an object identifier assigned to the object, and further associating the marker with the harvesting decision.

Accordingly, an unmanned vehicle is an unmanned aerial vehicle or an unmanned ground vehicle. The advantage of an airborne vehicle is its increased vertical range so that more detailed information can be obtained on how an object should be harvested. You can also negotiate areas where the ground is not suitable for land vehicles, such as crossing rivers and swamps. On the other hand, a ground vehicle may instead carry more equipment, such as sensors, and may have a longer battery life. The unmanned vehicle may be configured to be remotely controlled such that the control unit or human user can direct the unmanned vehicle to areas of interest as well as select objects to be analyzed. Alternatively, the driverless vehicle may be an autonomous vehicle.

Accordingly, the marker may be a physical marker, and the unmanned vehicle may be configured to associate the marker with an object by attaching a physical marker to an object or location in the immediate vicinity of the object, or by placing a physical marker on an object or location in the immediate vicinity of the object. Thus, an unmanned vehicle can place markers in forested areas so that the harvester or operator using the harvester can identify the markers and harvest objects based on them. The physical tag may suitably be a wireless tag, such as an RFID tag, wherein the wireless tag can be controlled to make a sound, release chemical(s), light up, emit smoke, etc. to facilitate recognition. Alternatively, the physical marker may be in the form of spray paint, paint, tape, plastic tape, staples, a sticker, a paintball, a cut mark or the like, or a chemical that is applied to the object. The position of the wireless tag can be obtained by triangulation or by determining the signal transmission power by known methods. The wireless tag can be configured to provide wireless radio messages when requested. Wireless radio messages can encode information such as: I am here, Don't come here, I am a log, I am a fallen log, I am a swamp, Come here, Cut me, Don't cut me and others.

The marker may also be a digital marker, and the autonomous vehicle may be configured to associate the marker with an object and a logging decision by storing the marker along with the object identifier and the logging decision in memory. The digital marker may be a reproduction of a bark pattern, a reproduction of a branch pattern/graphic tree identifier, or an image of a cut mark placed on the property by an unmanned vehicle.

A physical or digital marker can optionally be associated with a location on a 2D map, 3D map, 3D model; point cloud or other scaled representation of the forested area so that the location where the feature is located can be linked to a point on the map to help locate the feature during logging.

Accordingly, the at least one sensor of the autonomous vehicle may comprise a stereoscopic camera, a time-of-flight sensor, an image sensor, a chemical analyzer, a LIDAR sensor, or radar equipment.

The present invention also includes a logging preparation system that contains at least one unmanned vehicle according to the invention and which further comprises a remote control unit including processing circuitry and/or at least one additional unmanned vehicle according to the invention, when herein said remote control unit and/or additional unmanned vehicle is/are communicatively connected to the unmanned vehicle. Thus, the control unit can control at least one unmanned vehicle, but alternatively, a plurality of unmanned vehicles that navigate through a forested area. The processing circuitry may be located on each of the autonomous vehicles or

- 4 046164 may be located in the remote control unit so that the unmanned vehicle(s) communicate(s) with the remote control unit to receive harvesting decisions and store markers.

Thus, the remote control unit may be suitably configured to receive information associated with a plurality of markers from at least one unmanned vehicle and create or update a map based on the received information.

The present invention further includes another harvesting system that comprises at least one unmanned vehicle in accordance with the invention and that further comprises at least one harvester, each harvester being communicatively coupled to a processing circuit included in the at least one unmanned vehicle or communicatively associated with it. Thus, the complete harvesting preparation and harvesting operation can be performed by an unmanned vehicle receiving sensor information about objects in the forest area, associating object IDs with objects and tagging them so that they can be harvested in an optimized manner by the harvester. Accordingly, the unmanned vehicle first operates in the forest area, and the harvester follows it later and uses the markers associated with the objects when harvesting forests. This has the advantage that the harvesting is prepared before the harvester itself is used, so that the harvesting can be planned in advance. Alternatively, the driverless vehicle can work simultaneously with the harvester and link markers to objects that are soon to be harvested by the harvester. This has the advantage that logging is time efficient and does not require extensive preparation before logging can begin. It is also possible to send a self-driving vehicle in different directions in a forested area depending on which location the harvester operator deems most suitable for harvesting at that particular time. The harvester may be configured to recognize markers and logging decisions associated with the markers and perform one or more logging actions in a forested area based on said markers and logging decisions.

The present invention further includes a computer-readable storage medium storing instructions that, when executed by the system processing circuitry, cause the system to: obtain information associated with an object; assign an identifier, ID, to an object based on the received sensor information; receive a decision on logging based on the received sensor information or object identifier, ID assigned to the object; and associate the marker with the object and the harvesting decision.

The computer-readable storage medium may further store instructions that, when executed by the system processing circuitry, cause the system to execute the method according to any of the appended claims.

The effects and/or advantages presented in the present invention for embodiments of the method according to the first aspect may also be applied to corresponding embodiments of a computer-readable storage medium.

Many additional benefits and advantages of the present invention will be readily apparent to one skilled in the art from the detailed description below.

It should be noted that embodiments of the present invention apply to all possible combinations of features set forth in the claims.

Blueprints

The invention will now be described in more detail with reference to the accompanying drawings, in which: FIG. 1 illustrates a use case scenario for a forestry preparation solution according to one or more embodiments;

fig. 2 is a schematic overview view of a logging preparation system according to one or more embodiments;

fig. 3 is a schematic overview view of a logging preparation system according to one or more embodiments;

fig. 4a is a flow diagram of a method of preparation for logging according to one or more embodiments;

fig. 4b is a flow diagram showing sub-steps of a logging preparation method according to one or more embodiments;

fig. 5a shows an exemplary unmanned vehicle according to one or more embodiments;

fig. 5b shows an exemplary autonomous vehicle according to one or more embodiments;

fig. 6h-c illustrate the definition of a unique pattern associated with an object, according to one or more embodiments;

- 5 046164 fig. 7a, b illustrate the definition of a unique pattern associated with an object, according to one or more embodiments;

fig. 8 is a flow diagram of a method of preparation for logging according to one or more embodiments.

All drawings are schematic, not necessarily to scale, and generally show only those parts necessary to explain the respective embodiments, while other parts may be omitted or simply suggested. Any reference number appearing on more than one drawing refers to the same object or element in all drawings unless otherwise noted.

Detailed description

Introduction.

Aspects of the present invention will be described more fully below with reference to the accompanying drawings. However, the devices and method disclosed herein may be implemented in many different forms and should not be construed as limited to the aspects set forth herein. Like numbers on the drawings refer to like elements throughout.

The terminology used herein is intended only to describe specific aspects of the disclosure and is not intended to limit the invention. The use of the singular form herein does not exclude the plural form unless expressly indicated otherwise.

In the description, the term harvester should be understood as a forest harvesting machine.

As described in more detail in the prior art and summary of the invention, known solutions related to preparing a forested area for logging involve a human forester walking independently through a selected forested area and manually making assessments regarding the identification of objects such as trees, areas , not suitable for harvester movement, blocking objects, objects that should be excluded from harvesting, for example, for cultural heritage or biodiversity purposes, etc. Known solutions further include a human forester manually marking identified objects in a manner that a human harvester driver can recognize through the harvester's windshield, so that the human driver can make an on-the-spot logging decision related to the object, such as cutting down one tree without cut down another tree, do not enter the territory, avoid an object in your path, etc. For this to work correctly, the human driver must first be able to see the marker through the windshield, the marker must be placed in a suitable location to be viewed from that location and angle, and secondly, the human driver must be alert and drive slowly enough to manually register the marking, evaluate its meaning and respond to it by adjusting the path of the harvester, prepare for harvesting, or take other appropriate action depending on how the marking is to be interpreted.

As can be clearly understood, known manual methods of preparation for logging are labor-intensive and expensive processes. In addition, the determination processes largely depend on the skills and experience of the forester, as well as the human harvester driver. In addition, currently known manual solutions result in labeling that can be perceived as arbitrary and with large inconsistencies in terms of quality and consistency. In addition, a human forester cannot be expected to evaluate and mark every tree, swamp, etc. within the territory, further exacerbating the arbitrary nature of the result with large inconsistencies in terms of quality and consistency.

Inventors realized that this labor-intensive, expensive, inefficient, experience-dependent, and unpredictable process could be improved. Embodiments of the present invention are aimed at solving or at least alleviating the problem at hand through an at least partially automated harvest preparation solution in accordance with the embodiments presented herein.

Methods, devices, systems and associated storage media for preparation for logging will be described below with reference to FIGS. 1-8.

For purposes of the present invention, an object identifier is defined as a set of information to identify a given object so that it can be distinguished from other objects. An object identifier may be a unique identifier of an object that distinguishes the object from any other object, or, alternatively, it may be information that identifies the object as belonging to a group of objects that have at least one characteristic in common. This general characteristic may be the category to which the object belongs, such as a tree, boulder, or swamp, or a subcategory within one such category, such as a tree with a height greater than a given threshold or a tree of a given species. Alternatively, the general characteristic could be a logging decision appropriate for a site, such as a tree to be harvested, a tree not to be harvested, or an area or object to be avoided.

Pre-stored object identifier alternatives can be defined as a set of such categories or characteristics that are determined before preparation for harvesting, so that

- 6 046164 each object being evaluated can be determined to belong to one of the available alternatives.

A harvesting decision is defined as a decision about whether and how a property should be harvested. A logging decision may simply be a tree to harvest or a tree not to harvest, but it can also be more detailed and specify how the logging should be carried out.

A logging action is defined as an action taken on an object during logging in a forested area. The act of harvesting may involve harvesting an object, harvesting an object in a particular manner, or alternatively, it may involve leaving an object in place or avoiding an object while being at least a predetermined distance from the object. The harvesting action is accordingly performed based on the harvesting decision associated with the object.

A unique pattern is defined as an appearance of an object that is unique to that object. The unique pattern may be a color pattern representing the distribution of color on an object or part of an object, and/or a shape or size of an object or part of an object that is different from the shape of any other object in the forested area. A unique pattern may be a pattern representing one or more characteristics on the surface of an object that can be recognized through image processing and that together form a pattern that is different from the pattern of any other object in the forested area. In a non-limiting example where the object is a tree, such a unique template could be a bark pattern reproduction, a branch pattern reproduction/tree identifier graphic, a unique cut mark, etc.

A unique position is defined as the position of an object that is different from the positions of other objects in the forest region. Therefore, the unique position preferably has a level of accuracy (5 m, 2 m, 1 m or even less accuracy) such that the position of neighboring objects can be distinguished from each other.

Fig. 1 describes the invention in general form with a forested area containing a plurality of objects 110, which may be trees. The at least one unmanned vehicle 100 is configured to move through a forested area and receive sensor information associated with objects 110 to assign an object ID to the object based on the sensor information using processing circuitry that may be contained in the unmanned vehicle 100 or that , alternatively, may be accessible to the unmanned vehicle by being located at a remote control unit 300 that communicates with each of the unmanned vehicles 100. The at least one unmanned vehicle 110 is further configured to associate the marker 130 with the object 110 and the information received. sensor and/or an object ID assigned to object 110.

Based on the received sensor or object ID information, processing circuit 210 may be configured to make a harvesting decision and optionally associate a marker with the harvesting decision as well. Therefore, a harvesting decision may be made by the processing circuit 210 contained in the unmanned vehicle 110 in connection with obtaining, using at least one sensor of the unmanned vehicle, information associated with the object, that is, while the unmanned vehicle is moving under the forest canopy. . Alternatively, the harvesting decision may be made by processing circuitry 210 contained in the unmanned vehicle, harvester 150, or remote device at a later time. In some embodiments, the harvesting decision may be made by a human operator who uses the information and assigned object identifiers to make a decision for each object. The harvesting decision may be based on received sensor information or an object ID assigned to a single object, or received sensor information and/or object IDs assigned to multiple objects. In some embodiments, the unmanned vehicle 100 is a single unmanned vehicle configured to perform the steps for each of at least one object 110 in a forested area: obtaining, using at least one sensor 120 of the unmanned vehicle 100, information associated with object 110; assigning an object identifier (ID) to the object 110 based on the received sensor information using processing circuitry 210 contained in or available to the unmanned vehicle 100; obtaining a harvesting decision using processing circuit 210 based on the received sensor information or an object identifier (ID) assigned to object 110; and associating the marker 130 with the object 110 and the logging decision in one single instance while moving under the forest canopy of a forested area.

In other embodiments, the unmanned vehicle 100 is a single unmanned vehicle configured to perform one or more of these steps in a first case while driving under the forest canopy of a forest area, and the remaining steps in a second case while moving under the forest canopy of a forest area. .

- 7 046164

In yet other embodiments, an unmanned vehicle 100 is defined as a set of two or more unmanned vehicles 100, wherein the first unmanned vehicle 100 of the two or more unmanned vehicles 100 in the set is configured to perform one or more of these steps in the first case when moving under the forest canopy of a forest area, and the second unmanned vehicle 100 of the two or more unmanned vehicles 100 in the set is configured to perform the remaining steps in the second case when moving under the forest canopy of a forest area.

In the following, the invention will be described in more detail, and a number of exemplary embodiments of the invention will also be disclosed.

Device and system architecture.

Next, embodiments of the apparatus and system will be described with reference to FIGS. 2, 3, 5a and 5b.

Let's look first at FIG. 5a and 5b, which show two exemplary embodiments of an autonomous vehicle 100 according to the invention. In all embodiments herein, the unmanned vehicle 100 is configured to prepare for logging and includes: a body 101 configured to move through a forested area under a forest canopy; and at least one sensor 120 configured to receive information associated with the object 110; wherein the unmanned vehicle 100 includes or is communicatively coupled with a processing circuit 210 configured to assign an object identifier ID to the object 110 based on the received sensor information; and optionally making a harvesting decision based on the received sensor information or an object ID assigned to the object 110. The unmanned vehicle 100 is further configured to associate the marker 130 with the object 110 and the received sensor information or an object ID assigned to the object 110, and, optionally, decision on logging. The unmanned vehicle 100 may be an unmanned aerial vehicle, as shown in FIG. 5a, or an unmanned ground vehicle as shown in FIG. 5b. In various embodiments, the autonomous vehicle 100 may be remotely controlled or may be an autonomous vehicle.

In some embodiments, the marker 130 may be a physical marker, wherein the unmanned vehicle 100 is configured to associate the marker 130 with an object 110 by attaching a physical marker to the object 110 or a location in close proximity to the object 110 or by placing a physical marker on the object 110 or location in close proximity to the object 110. To achieve this, the unmanned vehicle 100 in these embodiments includes advantageous tagging equipment 102. Tagging equipment 102 includes suitable means for applying physical markers or wireless tags to objects. The physical tags may, for example, be one or more of the following: a radio frequency identification (RFID) tag or other wireless tag; spray paint or other paint; tape; plastic tape or paper tape; staples; sticker; a cut mark or the like that can be placed there by the unmanned vehicle 100 for later recognition; or a chemical substance that can be detected by the sensor of a chemical analyzer. A physical marker may be applied to an object by being dropped on or in close proximity to the object, by spraying or applying paint to the object, or by staples to the object, or by tying around the object.

In other embodiments, or in combination with the use of a physical token, the token 130 may be a digital token, wherein the unmanned vehicle 100 is configured to associate the token 130 with the object 110 and the received sensor information or object ID assigned to the object 110, and, optionally, with the logging decision by storing a marker 130 associated with the received sensor information or object ID and optionally the logging decision in memory 140. Some non-limiting examples of digital markers are a bark pattern reproduction, as shown in FIG. 6a-c, reproduction of the branch pattern/graphical identification of the tree, as shown in FIG. 7a-b, an image of a cut mark) or other physical marker (which could have been placed there by an unmanned vehicle using marking equipment 102), a color pattern, temperature, humidity and/or chemical composition associated with the object, or other unique feature or set features that can be used to identify the object 110.

A physical or digital marker can optionally be associated with a location on a 2D map, 3D map, 3D model; point cloud or other representation at the scale of the forest area.

At least one sensor 120 of the unmanned vehicle 100 may include one or more of the following: a stereoscopic camera; time-of-flight sensor; another image sensor; chemical analyzer; LIDAR sensor; or radar equipment.

In fig. 2 illustrates a logging preparation system 200 comprising at least one unmanned vehicle 100 in accordance with any of the embodiments disclosed herein, and further comprising a remote control unit 300 comprising

- 8 046164 processing circuit 210 and/or at least one additional unmanned vehicle 100, in which said remote control unit 300 and/or additional unmanned vehicle 100 are communicatively coupled with the unmanned vehicle 100. As a consequence, the method steps of any of embodiments of the method described in connection with FIG. 4a and 4b may be performed by the processing circuit 210 in the unmanned vehicle 100 or the additional unmanned vehicle 100, or by the processing circuit 210 in the remote control unit 300. Of course, some method steps may also be performed by processing circuitry 210 in the autonomous vehicle 100 or additional autonomous vehicle 100, while others are performed by processing circuitry 210 in the remote control unit 300, whichever is most appropriate. In some embodiments, the remote control unit 300 may be configured to receive information associated with a plurality of markers 130 from at least one autonomous vehicle 100 and create or update a map or other two-dimensional or three-dimensional representation of a forested area based on the received information.

As shown in the embodiments of FIGS. 3, the invention also includes a forest harvesting system 400 comprising at least one unmanned vehicle 100 according to any of the embodiments disclosed herein, and further comprising at least one harvester 150, wherein each harvester 150 is communicatively coupled to at least with at least one unmanned vehicle 100. Each harvester 150 of the logging system may be configured to recognize markers 130 and, if applicable, harvesting decisions associated with markers 130, and further perform one or more harvesting actions in the forested area based on specified markers 130 and logging decisions, if any, were made and saved. Alternatively, as described herein with reference to a later time, the harvesting decision may be made during harvesting by the processing circuitry 210 contained in the harvester, an accompanying unmanned vehicle, or at a remote location, or the operator makes the decision and enters the information into harvester.

Processing circuit 210 may, for example, comprise one or more processors. The processor(s) may be, for example, Application-Specific Integrated Circuits (ASICs) configured to perform a particular eye tracking and position detection method. Alternatively, the processor(s) may be configured to execute instructions (e.g., in the form of a computer program) stored in one or more memory devices 140. Such memory 140 may, for example, be contained within the system 200 or may be external to to the system 200 (for example, located remotely from it). Memory 140 may store instructions causing system 200 to perform a method according to any of the embodiments presented in connection with FIG. 4a, 4b and 8.

Processing circuit 210 may, in one or more embodiments, be configured to perform any or all of the method embodiments described in connection with FIG. 4a, 4b and 8.

It should be understood that the embodiments of the autonomous vehicle 100 and system 200 described above with reference to FIGS. 2, 3, 5a and 5b are provided as examples, and many other unmanned vehicles and systems may be envisaged.

Method embodiments

Next, embodiments of the method will be described in connection with FIG. 4a and 4b.

In fig. 4a shows embodiments of a method of preparing for logging using an unmanned vehicle 100 configured to move under a forest canopy in a forested area, the method including, for each of at least one object 110 in the forested area:

At step 410: obtaining, using at least one sensor 120 of the unmanned vehicle 100, information associated with the object 110. The information may be dimensions of the object, such as volume, height and/or surface area; the location of the object or the boundaries of its extreme points, the color of the object, the assigned class of the object, such as tree, swamp, boulder, etc., or alternatively, it could be the results of segmentation or classification of the object using machine learning.

In one non-limiting example, one or more of the at least one sensor 120 is configured to collect terrain information. For example, one or more of the at least one sensor 120 may be a lidar sensor configured to collect information to create a Digital Elevation Model (DEM), a Digital Surface Model (DSM), or a digital terrain model (DSM). Digital Terrain Models (DTM), Triangular Irregular Networks (TEST), or any other suitable representation.

Each of the at least one object may be, for example, any of the following: a tree or part of a tree, the boundary of an area, an area not suitable for harvester movement,

- 9 046164 a boulder existing in a forested area, a preferred means of movement, a site for collecting chopped logs, an ancient remains or monument, or a biological object of protection.

At step 420: assigning an object identifier (ID) to object 110 based on the received sensor information using processing circuit 210 contained in or available to the autonomous vehicle 100.

Assigning an object ID based on received sensor information may in various embodiments include, for example, processing 2D or 3D image data, measuring temperature or humidity (or detecting deviations from such measurements), and/or using chemical analysis.

In one or more embodiments, assigning an object identifier to an object 110 using processing circuit 210 includes analyzing sensor information to determine at least one property of the object 110 and selecting an object identifier from a set of pre-stored object identifier alternatives based on the at least one property of the object . Pre-stored object identifier alternatives may, for example, include identifiers indicating that object 110 is a tree to be cut, a tree not to be cut, an area the harvester should not enter, an object not to be approached, or any other suitable identifying information that is useful for harvesting purposes. The at least one object property may be an object type, for example, tree, tree species, swamp, river, fallen log; or alternatively it may be the color, surface roughness ratio, size, volume, shape or pattern of an object, or alternatively it may be a geographic location or interaction with another object, such as positioning relative to terrain, which may be vertical, horizontal or at a certain angle to any of these directions.

In other embodiments, assigning an object identifier to object 110 using processing circuit 210 includes: analyzing received sensor information to determine a unique pattern associated with object 110; generating a unique object identifier based on a certain unique template; and assigning a unique object identifier to the object 110. In some embodiments, the acquired sensor information comprises two-dimensional or three-dimensional image data or three-dimensional point cloud data, and a unique pattern is determined by processing the image data or point cloud data to recognize the pattern. The unique object identifier then provides a direct mapping between the real world object 110 and the corresponding object representation stored in memory 140.

In some embodiments, the corresponding object representation is associated with a unique position in the 2D or 3D representation of the forest region. According to a non-limiting example in which the object 110 is a tree, the image pattern determined based on the analysis of the acquired sensor information may be a reproduction of a bark pattern indicative of visible features of the tree's bark. This is shown in Fig. 6a-6c, where the bark of an object (tree) 110 contains a number of features 601 recognized from 2D or 3D image data or 3D point cloud data collected regarding the object 110 by at least one sensor 120 of the unmanned vehicle 100. By analyzing the sensor information, in In this case, 2D or 3D image data or 3D point cloud data containing features 601, processing circuit 210 is configured, and method step 420 includes, defining a template 602, as illustrated in FIG. 6b, where the detected cortical features 601 are combined to create a template 602. The template 602 shown without object 110 in FIG. 6c is unique to object 110 and is therefore suitable for use as a unique object identifier. According to another non-limiting example, in which the object 110 is a tree, the image pattern determined based on the analysis of the acquired sensor information may be a branch pattern reproduction, which may also be referred to as a tree pattern reproduction or a key graphical qualifier indicating the distribution of tree branches. This is shown in Fig. 7a and 7b, where a series of branch end features 701 are recognized from 2D or 3D image data or 3D point cloud data collected regarding the object 110 by at least one of the sensors 120 of the autonomous vehicle 100. By analyzing the sensor information, in this case 2D or 3D image data or 3D point cloud data containing features 701, processing circuit 210 is configured to, and method step 420 includes, define a template 702, as illustrated in FIG. 7b, where the detected tree features 701 are combined to create a template 702. The template 702 is unique to the object 110 and is therefore suitable for use as its unique object identifier. Another example of what may constitute a pattern in the context of step 420 is, for example, a color pattern associated with an object 110 recognized through analysis of two-dimensional or three-dimensional data acquired by at least one sensor 120 of the unmanned vehicle 100. Depending, of course, Depending on which sensor or sensors are deployed on the autonomous vehicle 100, other types of unique patterns may be created. In various embodiments, a certain unique

- 10 046164 The given pattern may be a pattern recognized directly by the sensors, or it may be a processed version of the pattern, as illustrated above.

In some embodiments, assigning an object identifier to object 110 using processing circuitry 210 includes analyzing sensor information to determine a unique position associated with object 110; generating a unique object identifier based on a certain unique position; and assigning a unique object identifier to object 110.

Of course, any combination of the embodiments described at step 420 may be used to generate an object identifier for object 110, and the object identifier may contain more than one type of information.

At an optional step 430: obtaining a harvesting decision using processing circuit 210 based on the received sensor information or an object identifier (ID) assigned to object 110.

In various embodiments, the harvesting decision may be a decision to harvest the item 110, how to harvest the item 110, not to harvest the item 110, or to avoid the location where the item 110 is located or a specific area in which the item 110 is located.

As illustrated in FIG. 4b, in some embodiments in which the object 110 is a tree to be cut, a method step 430 for obtaining a harvesting decision based on the received sensor information may include, and the processing circuit 210 may be configured at a substep 431: calculation by the circuit 210 processing the three-dimensional shape of a tree trunk based on the received sensor information;

at substep 432: the processing circuit 210 calculates the optimal method of cutting the tree trunk at at least one identified location to maximize the possible yield of high quality wood for the tree or log.

In these embodiments, associating marker 130 with object 110 and making a logging decision at step 440 includes, at substep 441: physically marking each of the at least one identified location on the tree trunk and/or digitally marking each of the at least one identified locations on the tree trunk in a digitally scaled 3D representation of the tree trunk.

The method, apparatus, and system in accordance with the embodiments presented herein may use at least one sensor 120 of the unmanned vehicle 100 and marking equipment 102 for a specific mark where the harvester 150 is to cut a tree or log, both at the base, and at specified intervals. up the trunk. This maximizes the possible yield of high-quality wood, which, for example, may correspond to the maximum amount of wood used and/or the cost of harvested wood for each tree cut. The maximum possible yield of quality wood for a particular tree depends on what it is intended for.

Optimizing the potential yield of high quality wood for a tree, as illustrated in sub-steps 431, 432 and 441, is commonly referred to as bucking.

At step 440: associating the token 130 with the object 110 and the received sensor information or object ID assigned to the object 110.

If step 430 has been completed, step 440 may further include associating marker 130 with a harvest decision. In some embodiments, step 440 may include associating the marker 130 with the object 110 and the logging decision only as the logging decision is based on, and therefore contains information about, the received sensor information, or an object ID assigned to the object 110.

To associate a token, you can store a digital token associated with a property or object ID in memory and/or apply a physical token to the object itself. The harvesting decision may be stored in a digital token or physical token, or alternatively may be stored in association with an object identifier. In one such example, the feature identifier might be that the feature is a swamp, and the associated logging decision might be to avoid the feature.

The marker 130 may be a physical marker, wherein associating the marker 130 with an object 110 involves attaching a physical marker to an object or location in close proximity to the object, or placing the marker 130 on or near the object 110.

The physical tags may, for example, be one or more of the following: an RFID tag with a radio frequency identification or other wireless tag; spray paint or other paint; ribbon; plastic tape or paper tape; staples; sticker; a cutting mark or the like that may be left there by the unmanned vehicle 100 for later recognition; or a chemical substance that can be detected by the sensor of a chemical analyzer. A chemical detector sensor can be installed on the harvester.

- 11 046164

Alternatively, or in combination with any of the above embodiments, the marker 130 may be a digital tag, wherein associating the marker 130 with the object 110 and the received sensor information or object ID assigned to the object 110, and, optionally, a harvesting decision includes storing the marker 130 along with the received sensor information or object identifier and, optionally, a harvesting decision in memory 140.

In some embodiments, the method further includes determining whether the mission has completed or whether there are still objects that need to be assessed. If it is determined that the mission has not completed or there is at least one more object to evaluate, the method returns to step 410. In other words, the method described in connection with FIG. 4a can be executed repeatedly while the logging mission is completed.

In some embodiments, the method further includes determining the remaining power supply of the unmanned vehicle and determining whether additional objects should be evaluated based on the power supply and, optionally, also based on the power required to return to a location where the unmanned vehicle can be recharged. .

The processing circuit 210 may suitably be configured to perform any of all of the method embodiments described in connection with steps 410-440.

In combination with any of the harvesting preparation embodiments presented herein, in which the marker 130 is a digital marker, the method may further include, and the processing circuitry may further be configured to store information associated with the at least one object 110, together with an object identifier 110 and an associated token 130 in a memory 140 accessible to the unmanned vehicle 100.

In combination with any of the embodiments presented herein to prepare for logging, the method may further include generating a 2D or 3D scale representation of at least a portion of the forest area or updating a 2D or 3D scale representation of at least a portion of the forest area by adding a marker associated with an object to a 2D or 3D view. The 2D or 3D representation may, for example, be a two-dimensional or three-dimensional map or model of a forest area containing location information.

Based on the information collected by the autonomous vehicle, the logging route can be determined so that the logging is carried out in an optimal manner. The logging route is based on information associated with at least the markers associated with the features, but may also be determined based on other information associated with the features in the forest area, such as their feature IDs, and preferably also based on scalable 2D or 3D -representing at least part of the forest area. Accordingly, the harvesting route is also determined based on at least one property of the harvester. In some embodiments, a logging route may also be determined based on establishing relationships between the locations of all the different objects and interactions between each object and the shape of the terrain, the distribution of rivers or nearby roads, etc.

In fig. 8 is a flowchart of a method according to one or more embodiments, in which steps 410 to 440 correspond to steps 410 to 440 described in connection with FIG. 4a, wherein the method further includes, at step 810: harvester 150 recognizing marker 130 using at least one sensor 160 of harvester 150;

at step 820: harvester 150 receives a harvesting decision for object 110.

In some embodiments, in which the resulting logging decision is a logging decision associated with the recognized marker 130. In other embodiments, obtaining a logging decision for object 110 involves processing circuit 210 or a manual operator generating a logging decision in any manner described herein. way.

At step 830: the harvester 150 performs a harvesting action based on the harvesting decision associated with the object 110.

Marker recognition may include digital image processing of data received from an image sensor, wherein the processing includes segmentation; feature recognition or color recognition, especially for species identification; or the recognition of chemical information using an electronic analyzer, in particular the recognition of methane to look for swamps or decaying vegetation.

Additional embodiments

In one or more embodiments, a computer-readable storage medium is provided storing instructions that, when executed by processing circuitry 210 of the system 200, cause the system 200 to perform a method as defined in any of the methods disclosed herein (in other words, in the claims, summary of the invention or detailed description).

The computer-readable storage medium may store instructions that, when executed by processing circuitry 210 of the system 200, cause the system 200 to obtain information associated with the object 110; at

- 12 046164 assign an identifier, ID, of an object to object 110 based on the received sensor information; optionally obtain a harvesting decision based on the received sensor information or an object ID assigned to object 110; and associate the marker 130 with the object 110, the received sensor information or object ID, and, if received, the harvesting decision.

In one or more embodiments, the computer-readable storage medium may further store instructions that, when executed by the processing circuit 210 of the system 200, cause the system 200 to perform steps or functions according to any of the method embodiments described herein.

A computer-readable storage medium may, for example, be provided in a computer program product. In other words, the computer program product may, for example, comprise a computer-readable storage medium storing instructions that, when executed by processing circuitry 210 of the system 200, cause the system 200 to perform a method as defined in any of the method embodiments.

As described above with reference to FIG. 2, the storage medium need not be contained in the system 200.

Example Embodiments

Next, six additional embodiments A-E will be described, which show various applications of the present invention in various preparation methods for logging. All embodiments are within the scope of the present invention.

A. Preparing for logging and applying physical markers in the forest area.

The at least one unmanned vehicle is guided through a forested area and is configured to obtain information relating to a plurality of objects, which are then marked with a number of physical markers that are visible to the harvester or the person operating the harvester during a subsequent harvesting operation.

The unmanned vehicle may be a drone (unmanned aerial vehicle) that flies through a forested area under the forest canopy, or may be a drone configured to move on the ground in a forested area. An autonomous vehicle can operate autonomously, but alternatively it can be controlled by a remote control or by an operator who controls the autonomous vehicle directly. In order for an autonomous vehicle to navigate a forested area, it is advisable for it to contain equipment to detect its surroundings, to navigate between trees and other objects, and to identify those objects that should be assessed and prepared for logging. Depending on what is appropriate for a particular logging operation that is to take place later, the autonomous vehicle can be configured to evaluate only those trees that are large enough to be of interest from a logging perspective, such as trees that are above a height threshold , such as 2 m, 4 m or 10 m, or trees whose maximum diameter exceeds a threshold, such as 20 cm or 30 cm. Alternatively, the driverless vehicle can also be configured to look at other objects near the trees, such as areas with water on the ground (such as rivers, ditches, or lakes) or areas with soft ground (such as swamps or bogs). This can be detected by an unmanned vehicle having sensors to detect humidity or having image processing equipment and capable of analyzing the images so that such areas can be identified based on the images captured by the unmanned vehicle.

Once in a forested area, the unmanned vehicle can evaluate each object that is determined to be suitable for assessment, for example, through identification in images captured by the unmanned vehicle, such as a drone, as described above. Each object is assessed using information related to the object received by the unmanned vehicle, or a sensor or sensors detecting information related to the object, or using information obtained through a connection with a remote control unit or the like. If the information is received from a remote control, it may be information that was previously collected and which, among other things, relates to the object itself or to the area where the object is located. If the autonomous vehicle itself captures information using at least one sensor, the information may be measurements related to an object, such as the height or width of a tree, the species the tree belongs to, which can be recognized using image recognition, or what - something else that is relevant when assessing an object for logging purposes.

Based on the information, the object is assigned an object identifier. In its simplest form, a self-driving vehicle can use predefined object identifier classes or alternatives (that is, object categories or classes) and place the object into one of them based on the information. Examples would be a tree suitable for logging, where the eligibility criteria for this object ID alternative could be a tree at least 3m in height,

- 13 046164 the tree has a maximum diameter of at least 20 cm or the tree is of the species 'Pinus sylvestris'. Any combination of criteria may be required for an object to fit into an object identifier alternative. Another example would be an area that should be avoided by the harvester, where the criteria for meeting this object ID alternative could be: the area has moisture above a threshold, or the ground cover is more than 50% covered by Dicranum scoparium, or the area contains rocks larger than a given size or even the area contains an object or set of objects that can be identified as ancient remains.

Alternatively, an object identifier may be assigned by defining a unique object identifier that uniquely identifies the object itself, such as by obtaining a reproduction of the tree's bark pattern or other characteristic that distinguishes it from other similar trees. Or, alternatively, the object identifier could be the exact location where the object is located; this may be determined by an autonomous vehicle containing equipment to accurately determine its location, or by a remote control capable of determining exactly where the autonomous vehicle is and sending that information to the autonomous vehicle.

Once an object ID is assigned, a harvesting decision can be made that determines whether and how the object should be harvested. In some cases, the harvesting decision is made by the driverless vehicle itself based on predetermined criteria, and in other cases, the remote control device makes the decision and transmits it to the driverless vehicle. In other cases, the harvesting decision may be made by a human operator and may be made after the site has been marked.

However, in most cases, the unmanned vehicle itself makes the decision about logging and associates the marker with the object. The driverless vehicle includes marking equipment for placing or attaching physical markers on or near an object, and this could be, for example, in the form of spray paint that the driverless vehicle sprays onto the object, or in the form of paper tape that is attached around object or directly on the surface of the object. The marker may be attached or adhered to the surface of an object or to a location in close proximity to the object, as defined herein. The marker in any embodiment described herein may further be removable. The unmanned vehicle 110 may further be configured to detach and collect the marker from the site after harvesting. Other examples as described above are also possible. If the markers are spray paint, they can be of different colors to indicate an object ID or harvesting decision so that it can be easily detected by the harvester or the operator operating the harvester. Once the desired number of objects has been assessed or the unmanned vehicle has moved through a specific forest area, preparation for logging is completed. In some examples, the unmanned vehicle is configured to continuously transmit its data to the control unit or other unmanned vehicles that form part of the same system, and in other examples, the data is transferred immediately after harvest preparation is completed, or the data is stored in a memory unit on the unmanned vehicle itself. vehicle.

B. Preparation for logging and applying digital markers in the forest area.

The at least one unmanned vehicle is guided through a forested area and is configured to obtain information relating to a plurality of objects, which are then tagged with a number of digital markers that are stored in a point cloud or in a two-dimensional or three-dimensional representation of the forested area and which are provided to a harvester or a person, to the harvester manager during the subsequent logging operation.

The unmanned vehicle may be a drone that flies through a forested area under the forest canopy, or may be a drone configured to move on the ground in a forested area. In the following, the terms unmanned vehicle and drone may be used interchangeably, and both should be interpreted in this context to mean any type of unmanned vehicle unless specifically stated otherwise. The drone can operate autonomously, but can alternatively be controlled by a remote control or by an operator who controls the drone directly. In order for a drone to navigate through forested terrain, it is advisable for it to contain equipment to detect its surroundings, to navigate between trees and other objects, and to identify those objects that should be assessed and prepared for logging. Depending on what is appropriate for a particular logging operation that is to take place later, the drone can be configured to evaluate only those trees that are large enough to be of interest from a logging standpoint, such as trees whose height exceeds a threshold such as 2m, 4m or 10m, or trees with a maximum diameter greater than a threshold, such as 20cm or 30cm. Alternatively, the drone can also be configured to look at other objects near trees, such as areas of water on the ground (such as rivers, ditches or lakes) or areas with soft soil (on

- 14 046164 example, swamps or bogs). This could be detected by a drone that has sensors to detect moisture or has image processing equipment and the ability to analyze the images so that such areas can be identified based on the images taken by the drone.

Once in a forested area, the unmanned vehicle can evaluate each object that is determined to be suitable for evaluation, for example, through identification in images captured by the drone, as described above. Each object is evaluated by object-related information received by the drone or by a sensor or sensors detecting information that relates to the object, or by information received through a connection with a remote control unit or the like. If the information is received from a remote control, it may be information that was previously collected and which, among other things, relates to the object itself or to the area where the object is located. If the drone itself captures information using at least one sensor, the information could be measurements related to the object, such as the height or width of the tree, the species the tree belongs to that can be recognized using image recognition, or something else , which is important when assessing a logging site.

Based on the information, the object is assigned an object identifier. In its simplest form, a drone can use predefined object identifier classes or alternatives (i.e. categories or classes of objects) and place the object into one of them based on the information. Examples would be a tree suitable for logging, where the criteria for meeting this object identifier alternative could be the tree is at least 3 m in height, the tree has a maximum diameter of at least 20 cm, or the tree is of the species Pirns sylvestris. Any combination of criteria may be required for an object to fit into an object identifier alternative. Another example could be an area that should be avoided by the harvester, where the criteria for meeting this object ID alternative could be an area with moisture above a threshold, or an area containing rocks larger than a given size, or even an area containing an object or set of objects that can be identified like ancient remains.

Alternatively, an object identifier may be assigned by defining a unique object identifier that uniquely identifies the object itself, for example by obtaining a reproduction of the bark pattern or other characteristics of the tree that distinguish it from other similar trees. Or, alternatively, the object identifier could be the exact location where the object is located; this can be determined by a drone containing equipment to pinpoint its location, or by a remote control capable of pinpointing where the drone is and sending that information to the drone. In some embodiments, the unmanned vehicle may include positioning equipment to determine location using GPS, GLONASS, accelerometer(s), radio equipment for radio triangulation, radio reference node, imaging device and access to image processing equipment for image matching or any other system for global or other positioning in the real world, or a combination thereof.

Once an object is assigned an object ID, a harvesting decision can be made that determines whether and how the object should be harvested. In some cases, the harvesting decision is made by the drone itself based on predetermined criteria, and in other cases, the remote control makes the decision and transmits it to the drone. In other cases, the harvesting decision may be made by a human operator and may be made after the object has been marked. In some embodiments, the harvesting decision may be made by a computer algorithm that considers all objects in the harvesting area and optimizes the harvesting route based on associations between each object and harvester type. Different harvesters have different parameters regarding how steep or swampy terrain they can negotiate, and based on this, the harvesting route can be determined accordingly.

However, in most cases, the drone itself makes the decision about logging and associates the marker with the object. Markers are digital markers that are stored as a series of data points in a point cloud or in a 2D or 3D representation of a forest area (i.e. a map). In some examples, the drone creates a map based on data points, but in other embodiments, the data points are used to update an already existing map so that it contains both a representation of the area and feature IDs and/or logging decisions made within that area.

In some examples, the drone transmits its data to the remote control where it is stored and a map is created or updated, but in other examples the data is stored in a memory unit on the drone itself. In some cases, multiple drones can operate independently or synchronize to assess different parts of a forest region and combine their data into a single map or point cloud.

C. Preparation for logging and installation of long-range wireless tags in a forest area.

At least one unmanned vehicle is guided through a forested area and is configured to obtain information related to a plurality of objects, which then

- 15 046164 are tagged with a series of long-range wireless tags that are visible to the harvester or the person operating the harvester during a subsequent harvesting operation.

The unmanned vehicle may be a drone that flies through a forested area under the forest canopy, or may be a drone configured to move on the ground in a forested area. The drone can operate autonomously, but can alternatively be controlled by a remote control or by an operator who controls the drone directly. In order for a drone to navigate through forested terrain, it is advisable for it to contain equipment to detect its surroundings, to navigate between trees and other objects, and to identify those objects that should be assessed and prepared for logging. Depending on what is appropriate for a particular logging operation that is to take place later, the drone can be configured to evaluate only those trees that are large enough to be of interest from a logging perspective, such as trees whose height exceeds a threshold such as as 2m, 4m or 10m, or trees with a maximum diameter greater than a threshold such as 20cm or 30cm. Alternatively, the drone can also be configured to look at other objects near trees, such as areas with water on the ground (such as rivers, ditches or lakes) or areas with soft soil (such as swamps or bogs). This can be detected by a drone that has sensors to detect humidity or that has image processing equipment and can analyze the images so that such areas can be identified based on the images taken by the drone.

Once in a forested area, the unmanned vehicle can evaluate each object that is determined to be suitable for evaluation, for example, through identification in images captured by the drone, as described above. Each object is evaluated by object-related information received by the drone or by a sensor or sensors detecting object-related information, or by information received through a connection with a remote control unit or the like. If the information is received from a remote control, it may be information that was previously collected and which, among other things, relates to the object itself or to the area where the object is located. If the drone itself captures information using at least one sensor, the information could be measurements related to the object, such as the height or width of the tree, the species the tree belongs to that can be recognized using image recognition, or something else , which is important when assessing a logging site.

Based on the information, the object is assigned an identifier. In its simplest form, a drone can use predefined object identifier classes or alternatives (i.e. categories or classes of objects) and place the object into one of them based on the information. Examples would be a tree suitable for logging, where the criteria for meeting this object ID alternative could be the tree is at least 3 m in height, the tree has a maximum diameter of at least 20 cm, or the tree is of the species Pirns sylvestris. Any combination of criteria may be required for an object to fit into an object identifier alternative. Another example could be an area that should be avoided by the harvester, where the criteria for meeting this object identifier alternative could be an area with moisture above a threshold, or an area containing rocks larger than a given size, or even an area containing an object or set of objects that can be identified like ancient remains.

Alternatively, an object identifier may be assigned by defining a unique object identifier that uniquely identifies the object itself, for example by obtaining a reproduction of the bark pattern or other characteristics of the tree that distinguish it from other similar trees. Or, alternatively, the object identifier may be the exact location where the object is located; this can be determined by a drone containing equipment to pinpoint its location, or by a remote control capable of pinpointing where the drone is and sending that information to the drone.

Once a site is assigned a site ID, a logging decision can be made that determines whether and how to harvest, or an optimized route to harvest the entire area. In some cases, the harvesting decision is made by the drone itself based on predetermined criteria, and in other cases, the remote control makes the decision and transmits it to the drone. In other cases, the harvesting decision may be made by a human operator and may be made after the object has been marked.

However, in most cases, the drone itself makes the decision about logging and associates the marker with the object. In this example, the markers are long-range wireless tags, such as RFID tags, that are applied by being released from a drone so that they touch the object they are marking. For example, a drone could carry a cannon or similar and shoot tags at the surface of a tree or into the ground. Each tag contains components to carry a signal that either uniquely identifies the tag or identifies the tag as belonging to a particular group or tags. In addition, the mark may contain identifiers that serve to make the mark more visible to the harvester or operator

- 16 046164 harvester, for example, the tag may light up, make a sound, or emit smoke or a chemical signal when harvesting.

What is described in Examples A, B, and C above generally reveals similar methods for evaluating multiple items in preparation for harvesting in a forested area, but differs in the way the items are labeled. It is also possible within the scope of the present invention to combine different types of markers, such that the drone applies physical markers or wireless tags to objects, and also generates digital markers that are stored or updated in a database, point cloud, or area map. Additionally, the drone may be configured to select physical markers or wireless tags for each object depending on the object ID or information related to the object, so that, for example, spray paint can be used for trees and wireless tags for areas to be targeted. avoid.

D. Preparation for logging and marking methods for optimal tree cutting.

The at least one unmanned vehicle is guided through a forested area and is configured to obtain information relating to a plurality of objects, which are then tagged so that a harvester or person operating the harvester in a subsequent harvesting operation can harvest the objects in a predetermined manner.

The unmanned vehicle may be a drone that flies through a forested area under the forest canopy, or may be a drone configured to move on the ground in a forested area. The drone can operate autonomously, but can alternatively be controlled by a remote control or by an operator who controls the drone directly. In order for a drone to navigate a forested area, it is advisable for it to contain equipment to detect its surroundings, to navigate between trees and other objects, and to identify those objects that should be assessed and prepared for logging.

Depending on what is appropriate for a particular logging operation that is to take place later, the drone can be configured to evaluate only those trees that are large enough to be of interest from a logging standpoint, such as trees whose height exceeds a threshold such as 2 m, 4 m or 10 m, or trees with a maximum diameter exceeding a threshold, such as 20 cm or 30 cm.

Once in a forested area, the unmanned vehicle can evaluate each object that is determined to be suitable for evaluation, for example, through identification in images captured by the drone, as described above. For each object that is identified as a tree suitable for logging, the drone uses at least one sensor to obtain information about the object and determines the optimal way to harvest that particular tree (i.e., bucking). Preferably, the sensor contains imaging equipment so that images of the object can be captured.

Based on the images and possibly other information collected by the sensors, the three-dimensional shape of the tree trunk is calculated. It is advisable to capture a series of images from different positions in relation to the tree to make it easier to determine the three-dimensional shape. The optimal method of sawing the tree trunk is then determined to maximize the possible yield of high quality wood for the tree or log based on it. The drone may contain predefined criteria for which wood shapes are considered desirable, such as logs of a certain length or logs with a uniform circumference over a given portion of their length. Additionally, depending on where the branches are located on the tree, the criteria may include cutting the tree in different ways to distribute parts in the log where the branches start in a certain way, or based on the three-dimensional shape of the trunk along which the tree can be cut to maximize output regardless of other criteria. In this way, the drone identifies at least one specific location where the tree should be cut down in order to harvest the tree in the best possible way, taking into account certain criteria or simply to maximize yield.

As in other embodiments, the drone associates an object identifier with the tree, and this is preferably a unique object identifier so that the individual information collected for the tree is associated with a unique object identifier such that details of how the tree is to be harvested are can be used by the harvester or by the operator operating the harvester later. In this example, a logging decision is individual information about how a tree should be optimally cut down.

The drone also associates at least one marker with the tree. This could be a physical marker, such as spray paint from a paint gun or similar, included in the marking equipment on the drone, or a cut in the bark made with a knife or similar, which is also included in the marking equipment. A physical marker is applied to at least one identified location on the tree trunk, preferably where a cut is to begin or end when the tree is harvested. For more complex tree harvesting, multiple locations on the trunk can be marked in this way. This way, the drone can mark exactly where the harvester needs to cut down a tree or log, both at the base and at precise intervals up the trunk. This

- 17 046164 maximizes the possible yield of high-quality wood, which, for example, can correspond to the maximum amount of wood used and/or the cost of cut wood for each tree cut. The maximum possible yield of quality wood for a particular tree depends on what it is intended for.

The drone can also associate a digital marker with a tree, and this can be in the form of a point or area in a digital and scalable 3D representation of the tree trunk, which can be stored in the drone or transferred to the control unit for storage in a memory unit. The drone was also able to associate both physical and digital markers with each object.

Once the drone has assessed a suitable number of objects that are trees suitable for harvesting, or once the drone has passed through a forested area to be harvested, digital markers indicating the optimal harvest of each tree can be used by the harvester or the operator operating the harvester to harvesting of trees that were assessed. Digital markers can be used as a series of markers that are identified as they match the trees they correspond to, but they can also be represented on a map or model of a forest area to make it easier to find each specific tree.

This example of how to prepare for logging can be advantageously combined with any of the examples A-C disclosed above. For example, any of Examples A through C can be used to identify a property suitable for logging, and what is said above with reference to Example D can be used to plan the optimal harvest of those properties once they have been identified.

E. Preparing for logging using multiple sensors to determine the location of objects.

In this example, which can be combined with any of the examples A-D described above, the autonomous vehicle is used to detect and locate objects using other types of sensors to determine properties that are not image-based and that are not provided by the remote control unit. having access to previously obtained information.

Such other types of sensors may be sensors configured to, for example, detect chemical compounds, humidity, temperature and light levels of the environment. This is suitable for more accurate detection of a swamp or swamp by detecting any of a number of chemicals that are commonly emitted from such areas. This can also be used to determine soil moisture to identify areas that are too wet and unstable for the harvester, or areas that are at a different temperature than the surrounding environment. This may indicate areas that are still frozen early in the year or that have started to freeze in the fall, making it easier or more difficult to operate the harvester there than at other times. Running water may also be localized at a lower temperature than surrounding areas. By determining the ambient light, it is possible to estimate how much vegetation is present on the ground or how dense the forest is in the area, and therefore how easy or difficult it is to get there with a harvester.

Thus, when using a drone to prepare for logging in this example, the drone moves through a forested area and collects sensor information related to features such as areas of forest that are not considered trees. Such sensors can also collect some information related to trees, although they are primarily used for localization and correct soil identification and assessment to determine a suitable logging operation in the region.

One skilled in the art will understand that the present invention is in no way limited to the preferred embodiments and examples described above. On the contrary, many modifications and variations are possible within the scope of protection provided by the appended claims. For example, the embodiments described above with reference to FIGS. 3 may, as explained herein, be performed in different orders and/or combined with additional method steps to form additional embodiments. Additionally, it should be appreciated that the system 200 shown in FIG. 2 is considered as an example, and that other systems may also perform the methods described above with reference to FIG. 3.

It should be appreciated that processing circuitry (or processor) 210 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application specific integrated circuit, programmable gate array, or any other suitable computing device, resource or combinations of hardware, software and/or coded logic capable of providing the functionality of a computer, either alone or in combination with other computer components (such as memory or storage media).

It will also be understood that a memory or storage medium (or computer readable medium) may comprise any form of volatile or non-volatile computer readable memory, including, but not limited to, persistent storage, solid state memory, remotely mounted memory, magnet

- 18 046164 digital media, optical media, random access memory (RAM), read-only memory (ROM), storage media (for example, hard drive), removable storage media (for example, flash drive, compact -compact disk (CD) or digital video disk (Digital Video Disk (DVD)) and/or any other volatile or nonvolatile, non-transitory devices, readable and/or computer-executable memory devices that store information, data and/ or instructions that can be used by the processor or processing circuits.

Moreover, modifications to the disclosed embodiments can be understood and implemented by those skilled in the art through practice of the claimed invention based on examination of the drawings, description, and appended claims. In the claims, the word containing does not exclude other elements or steps, and the use of a feature in the singular does not exclude plurality. In the claims, the word or should not be interpreted as exclusive or (sometimes called XOR). In contrast, expressions such as A or B cover all cases of A and not B, B and not A, and A and B, unless otherwise specified. The fact that certain measures are set forth in mutually different dependent claims does not mean that a combination of these measures cannot be used to advantage. Any reference numerals in the claims should not be construed as a limitation of scope.

It should also be noted that features from the various embodiments described herein may be freely combined unless it is expressly stated that such a combination would be unsuitable.

Claims (21)

1. A method of preparing for logging in a forest area, including receiving, using at least one sensor (120) of an unmanned vehicle (100), configured to move under the forest canopy in the said area, information related to objects (110) located in the said area; assigning a marker (130) to each of said objects (110) in accordance with the received information about this object; obtaining a logging decision using a processing circuit (210) based on the received sensor information; associating marker (130) with a logging decision; recognizing said marker (130) using at least one sensor (160) of the harvester (150); performing a harvesting action with the harvester (150) in accordance with the recognized marker (130). 2. The method of claim 1, wherein assigning a marker (130) to each of the objects (110) using the processing circuit (210) includes analyzing sensor information to determine at least one property of the object (110) and selecting an identifier (ID) ) of an object among a set of pre-stored object identifier alternatives based on at least one property of the object (110). 3. The method of claim 1, wherein assigning a token (130) to each of the objects (110) using the processing circuit (210) includes: analyzing sensor information to determine a unique pattern associated with an object (110); generating a unique identifier (ID) of an object based on a certain unique template; assigning an identifier (ID) to an object (110). 4. The method of claim 1, wherein assigning a marker (130) to each of the objects (110) using the processing circuit (210) includes analyzing sensor information to determine a unique position associated with the object (110); generating a unique object identifier (ID) based on a certain unique position and assigning an identifier (ID) to the object (110). 5. The method according to any one of claims 3 or 4, further including storing information associated with at least one object (110), along with an identifier (ID) of the object and the associated token (130) in a memory (140) accessible for an unmanned vehicle (100). 6. The method as claimed in any one of the preceding claims, wherein the action to be performed by the harvester (150) includes one of the following: harvesting the object (110), harvesting the object (110) in a predetermined manner, avoiding the object, avoiding the place where the object (110) is located, or a certain area in which the object (110) is located. 7. The method of any one of the preceding claims, further comprising generating a scaled two-dimensional (2D) or three-dimensional (3D) representation of at least a portion of the forested area, or updated a scaled two-dimensional or three-dimensional representation of at least a portion of the forested area by adding a marker, associated with an object (110) in 2D or 3D representation. - 19 046164 8. The method of any one of the preceding claims, wherein the token (130) is a physical token and wherein the assignment of the token (130) to an object (110) involves attaching a physical token to the object (110) or at a location in close proximity to the object (110). ) or placing a marker (130) on or near an object (110). 9. The method according to any one of claims 1 to 7, wherein the marker (130) is a digital marker and wherein assigning the marker (130) to the object (110) includes storing the marker (130) along with the harvesting decision and the location of the object ( 110) in memory (140). 10. The method as claimed in any one of the preceding claims, wherein the object (110) is a tree to be cut, and wherein the action performed by the harvester (150) in accordance with the recognized marker (130) includes the circuit (210) calculating processing the three-dimensional shape of the tree trunk based on the received sensor information and calculating by the processing circuit (210) an optimal method for cutting the tree trunk at at least one identified location to maximize the possible yield of high quality wood for the tree; wherein assigning a marker (130) to an object (110) includes physically marking each of the at least one identified location on the tree trunk or digitally marking each of the at least one identified location on the tree trunk in a digitally scaled 3D representation of the tree trunk . 11. The method of any one of the preceding claims, further comprising generating a logging route based on a plurality of markers (130) assigned to objects (110) and a two-dimensional or three-dimensional scale representation of at least a portion of the forested area. 12. An unmanned vehicle (100) configured to prepare for logging in a forested area, comprising a body (101) configured to move through a forested area under a forest canopy; at least one sensor (120) configured to receive information associated with objects (110) located in said area; wherein the unmanned vehicle (100) is configured to assign a marker (130) to each of said objects (110) in accordance with the received sensor information about this object, the unmanned vehicle (100) contains or is communicatively connected with a processing circuit (210), configured to make or receive a harvesting decision based on the received sensor information, the unmanned vehicle (100) is further configured to associate a marker (130) with the harvesting decision, and each of said markers is configured to be recognized by the harvester (150) using at least at least one sensor (160) of the harvester (150) and to ensure that the harvester (150) performs a harvesting action in accordance with the recognized marker (130). 13. The unmanned vehicle of claim 12, wherein the token (130) is a physical token and wherein the unmanned vehicle (100) is configured to assign the token (130) to an object (110) by attaching the physical token to the object (110) or at a location in close proximity to the object (110), or by placing a physical marker on the object (110) or at a location in close proximity to the object (110). 14. The unmanned vehicle according to claim 12, wherein the token (130) is a digital token and wherein the unmanned vehicle (100) is configured to assign the token (130) to the object (110) by storing the token (130) along with the decision to logging and the location of the object (110) in memory (140). 15. The unmanned vehicle of claim 14, wherein the digital marker is a bark reproduction, a branch reproduction/graphic tree identifier, or a cut mark image. 16. An unmanned vehicle according to any one of claims 12-15, in which a physical or digital marker is further associated with a location on a two-dimensional map, three-dimensional map, three-dimensional model; point cloud or other representation at the scale of the forest area. 17. System (200) for preparation for logging, containing at least one unmanned vehicle (100) according to any one of claims 12-16 and additionally containing a remote control unit (300) containing a processing circuit (210) and/or at least one additional unmanned vehicle (100) according to any one of claims 12 to 16, wherein said remote control unit (300) and/or additional unmanned vehicle (100) is communicatively connected/connected with the unmanned vehicle (100 ). 18. The system (200) according to claim 17, in which the remote control unit (300) is configured to receive information associated with a plurality of markers (130) from at least one unmanned vehicle (100) and create or update a map on based on the information received. - 20 046164 19. System (400) for logging, containing at least one unmanned vehicle (100) according to any one of claims 12-16 and additionally containing at least one harvester (150), with each harvester (150) communicatively connected with a processing circuit (210) that is contained in or communicatively coupled with at least one unmanned vehicle (100). 20. The system (400) of claim 19, wherein each harvester (150) is configured to recognize markers (130) and harvesting decisions associated with the markers (130), or make harvesting decisions based on the markers and make one or more more logging activities in a forested area based on mentioned markers (130) and logging decisions. 21. A computer-readable storage medium storing commands that, when executed by the processing circuit (210) of the system (200) for preparation for logging according to claim 17 or 18, cause the system (200) to execute the method according to any one of claims. 1-11.