WO2023180758A2

WO2023180758A2 - Autonomous deposition machine with improved deposition layout features

Info

Publication number: WO2023180758A2
Application number: PCT/GB2023/050751
Authority: WO
Inventors: Anthony David George Rhoades; Samuel Paul CORNISH-EVANS; Yao GONG
Original assignee: Micropply Limited
Priority date: 2022-03-24
Filing date: 2023-03-24
Publication date: 2023-09-28
Also published as: GB2616895A; GB202211532D0; GB202204188D0; WO2023180758A3

Abstract

An autonomous deposition machine and associated method comprising at least one receptacle to hold a deposition material; at least one deposition arrangement; a locomotion arrangement; a sensor arrangement; a control unit, the control unit operable to receive at least one deposition instruction; wherein the deposition instructions include instructions for a multi section deposition; the control unit controlling the least one deposition arrangement and the locomotion arrangement to autonomously deposit the deposition material onto a surface in a first section, wherein during that first section, a true path is also deposited by the autonomous deposition apparatus, and wherein in a second section of the deposition, the control unit uses the sensor arrangement to dynamically monitor the true path and correct the position of the autonomous vehicle in order to deposit the second section in line with the true path. Thus, there is provided an improved high-resolution grand-scale accuracy of ground printing and deposition systems. Furthermore, delivering world leading navigational accuracy for a ground marking system ensuring market-leading flexibility, scalability, ease-of-use, and 20 robustness for the ground marking systems.

Description

AUTONOMOUS DEPOSITION MACHINE WITH IMPROVED DEPOSITION LAYOUT FEATURES

The present invention relates to the use of a sensor and deposition system to improve the print accuracy and scalability, for an Autonomous Distributed Deposition Robot (ADDR). An ADDR of a type equipped to deposit materials such as an ink and paint, but which may equally deposit sand, seed, fertiliser, or other ground treatments onto a ground surface, or for injection under pressure into a ground surface. Each ADDR may be completely autonomous (i.e. free from human operation and/or supervision) or may require at least partial human operation and/or supervision depending on the application.

Ground marking is typically carried out manually. It requires significant pre-planning, the manufacture of pre-ordered plastic stencils, and large teams of workers to decipher instructions, prepare, lay out, and complete a site for marking. Where marking is required such as for logos, safety or hazard signs, the complex make-up of these images mean that difficulties persist to print any image, any size, any colour, directly onto any ground surface without significant cost of time, expense and compromise in image attributes such as resolution.

Inaccuracies within a print, are more important than the inaccuracy of the overall print in terms of localisation. The human eye can see even very small inaccuracies within an image but tends to be less concerned about the accurate rotation and orientation of that image overall. One approach to scalable autonomous ground marking is found in the Applicant's co pending patent "Ground Printing Machine", Micropply Limited, PCT/GB2021/052671, which discloses an ADDR machine capable of ground printing and which uses the tiling of segments to cover an image print area.

Another approach is found in Pixelrunner's application US2019381529, which disclose using a camera and computer vision to compensate for a degree of adjustment between print segments, "Thus, in this example the actual progression of the second path segment may deviate from the theoretical target progression as the robot trajectory can be continuously corrected while moving through the second path segment to adapt the robot trajectory of the second path segment (and thereby the second section of the image) to the first image section."

Other autonomous vehicle systems may follow a pre-determined path, path follow another machine, or follow a virtual path (eg SLAM), however there is a need to lay down a path in relation to a section of a deposition, in order to maintain zero discrepancy in the x, y and z directions for a subsequent section of that deposition, in relation to the preceding section of deposition or ground marking.

However where a print or deposition has section of zero marking or no deposition, it can be difficult to follow the edge of an existing print/deposition, or follow a virtual path that is accurate enough.

Summary of Invention

According to a first aspect of the present invention, there is provided a ground deposition and/or ground marking autonomous robot apparatus and method comprising at least one receptacle to hold a deposition material; at least one deposition arrangement; a locomotion arrangement; a sensor arrangement; a control unit, the control unit operable to receive at least one deposition instruction; wherein the deposition instructions include instructions for a multi section deposition; the control unit controlling the least one deposition arrangement and the locomotion arrangement to autonomously deposit the deposition material onto a surface in a first section, wherein during that first section, a true path is also deposited by the autonomous deposition apparatus, and wherein in a second section of the deposition, the control unit uses the sensor arrangement to dynamically monitor the true path and correct the position of the autonomous vehicle in order to deposit the second section in line with the true path.

Thus advantageously, an accuracy of tiling alignment of +/-2mm and scalable to 50m² activations can be achieved.

Preferably the sensor system comprises a computer vision system, the computer vision system comprising a camera. Further preferably wherein the true path is a code or a pattern. Advantageously the code may be a QR code and/or a repeating pattern, and/or a repeating pattern of dots and spaces. Preferably, the pattern and/or code can be seen in its entirety within the field of view of the computer vision system.

Further preferably, comprising dynamically determining an offset with regard to the true path and operating one or more locomotion arrangements to minimize said offset. And/or further comprising dynamically determining an angular offset of the rotation angle (b) between the true path and the apparatus and the control unit controlling one or more locomotion arrangements to minimize said angular offset.

Preferably wherein the locomotion arrangement is a ground wheel arrangement and/or the apparatus further comprises a chassis with a nozzle array on a traverse guide, the traverse guide permitting movement of the nozzle array beyond the width of the ground wheel arrangement.

Wherein the deposition instructions may be a command to print an image in a certain size and the control unit calculates the required sections of the print. In use, the robot is able to move forward in a series of discrete steps (ranging in size from a minimum of 1mm up to 100's of mm), so the scale of movement and thus path corrections can be very small, almost to a blade of grass accuracy.

Advantageously the robot is configured with a nozzle array to print an image or logo on a surface, the robot housing two or more flexible bags containing a material for deposition, the material for deposition contained within each flexible bag being an ink or paint selected from a cyan, magenta, yellow, black, white, green, blue, or red colour, the image or logo optionally being an advertising logo, design or safety warning. Alternatively, wherein the material for deposition is a herbicide, pesticide, insecticide, plant growth aid, water or marking material, optionally wherein the marking material is a paint, ink, coloured material, powder.

Preferably, when the robot is in use depositing material on the ground, the onboard control system is configured to periodically gather data from the ADDR. The onboard control system is configured to transmit data to a remote resource, such as a cloud server, or an edge device optionally a tablet or smartphone.

Accordingly, there is provided a ground marking autonomous robot that in addition to high accuracy and throughput marking provides for robot diagnostics, data aggregation and secure communication links between the edge, the cloud and all data processing devices as required. The ground marking autonomous robot is combined with artificial intelligence, machine learning, and an end-to-end Cloud SAAS (Software as a Service) platform that work together to create ground-printed images approaching the accuracy of a blade of grass, all underpinned with an advanced user-interface.

Thus, there is provided an improved high-resolution grand-scale accuracy of ground printing and deposition systems. Furthermore, delivering world leading navigational accuracy for a ground marking system ensuring market-leading flexibility, scalability, ease-of-use, and robustness for the ground marking systems. With these elements in place, ADDRs such as the one disclosed in this application are able to fully satisfy even the most extreme scale market demands such as 'full pitch' print activations used in the NFL (National Football League).

Figures

Embodiments of the invention will now be described by way of example only and with reference to the accompanying drawings, in which:

Figure 1 is a schematic diagram of a robot comprising an array of primary packaging comprising bags filled with a ground marking material;

Figure 2 is a schematic diagram of primary packaging comprising a flexible ink bag with a hose connected to a nozzle array;

Figures 3a and 3b are plan views of a ground marking robot;

Figure 4 is a side elevation of a ground marking robot;

Figure 5 is a plan view of a ground marking operation in progress, in this embodiment tiled printing of an image;

Figure 6 is a diagram of a simple image which is difficult to print in a tiled manner. Figure 7 is a diagram of a sequence of dots with a specific pattern and an example camera view, according to one embodiment of the present invention;

Figure s is a process flow diagrams of the true path solution, according to one embodiment of the present invention.

Figure 9 is a schematic diagram of a smart communications module as used in the robot; and

Figure 10 is a schematic diagram of a secure communications network between the robot, the edge, the cloud and a data processing device.

The present techniques will be described more fully hereinafter with reference to the accompanying drawings. Like numbers refer to like elements throughout. Parts of the autonomous ground printer are not necessarily to scale and may just be representative of components of the ground print machines, or other described entities.

Detailed Description

Referring to Figure 1 a schematic diagram of an autonomous ground marking robot 10 comprises an outer case 12 cut away to reveal an array of primary packaging 14, 16, 18 and 20. The primary packaging 14, 16, 18 and 20 shown here comprises ink held within a bag (not shown in Figure 1), with primary packaging 14 comprising a red ink R, a green ink G, a blue ink B and a white ink W. Each primary packaging 14, 16, 18 and 20 is supported on a weight measuring plate 14a, 16a, 18a and 20a connected to a smart communications module 22 described more fully in Figures 9 & 10, which may also serve as or be connected to an onboard control system (not shown in Figure 1). The smart communications module 22 comprises a transceiver 22a for communication with remote resources (not shown in Figure 1).

Each weight measuring plate 14a, 16a, 18a and 20a is an integral part of a frame 26 capable of holding the primary packaging 14, 16, 18, 20 firmly in place and comprises a load sensor 28 for registering the presence of the primary packaging 14, 16, 18, 20 when firmly in place in the frame 26. Load sensor 28 may be a photodiode or a RFID tag that communicates with an ID tag 30 of the primary packaging 14, 16, 18, 20. ID tag 30 may also comprise a barcode or other smart label, which is used for identification of the primary packaging 14, 16, 18, 20.

More than one load sensor 28 can be used for load balancing. For example, two, three, four or more load sensors can be positioned as part of or under a platform or frame (which may, for example, be the weight measuring plate 14a) supporting the ink bag and primary packaging. In operation, when the platform is flat then all the load sensors should measure the same normal force to each load sensor 28 i.e. the force in line with their mounting and perpendicular to the flat surface or platform supporting the ink bag and primary packaging. When the robot is on an incline then the platform is on an incline and so the load of the ink in the ink bag is not distributed evenly across the platform and the load sensors will show different readings. As gravity acts perpendicular to a 0° incline any deviation from this horizontal must be accounted for in any measurements. The robot determines it is at an angle or incline from an onboard accelerometer and can report any incline in 3-axes. To account for the incline, trigonometry can be used to convert the normal force the scale measures into the weight. This calculation can be done with the 3-axis vector (incline in 3-axes) extracted from the accelerometer. Therefore, the method includes reading an output of load sensors and applying a corrective formula and adjusting weight measurement to take into account the incline and determine an accurate weight and/or volume.

As best seen in Figure 2, a flexible ink bag 32 comprises an airtight valve outlet 34 sealed to the flexible ink bag 32 with the appropriate connection part for secure connection to a hose 36. The hose 36 may also be a tube, piping or any suitable means to transport the material for deposition.

The autonomous ground marking robot 10 comprises wheels 24 for movement, a position sensor 38 and laser 40. Position sensor 38 may comprises a Global Positioning Device for navigation or the autonomous ground marking robot 10 may use triangulation with known positioning reflectors and the laser 40 for positioning. In operation, the robot may be in constant communication with a positioning device and may reposition itself based on communication from a Global Positioning Device, for example to perform turns, or to decide to return to a base unit for a fill up of deposition materials, or to return to a base at the end of a deposition cycle.

Turning to Figure 2, the primary packaging 14 comprising the flexible ink bag 32 with the hose 36 is connected to a nozzle array 42 via an actuator pump 35. Here the nozzle array 42 acts as the means to deposit the material for deposition. Any such suitable nozzle, nozzle array or means to deposit the material, depending on the actual material to be deposited, may be used. Each ink bag of the primary packaging 14, 16, 18 and 20 of Figure 1 will have a hose 36 and valve 34 to connect to the nozzle array 42 via the actuator pump 35.

The system may have a single actuator pump 35 for all primary packaging/ink bag/hose (14,16,18,20/32/36), or there may be multiple actuator pumps, i.e. one for each primary packaging/ink bags/hose (14,16,18,20/32/36). Each nozzle of the nozzle array 42 may be designated for each primary packaging/ink bag/hose (14,16,18,20/32/36) present, so that each nozzle is for deposition of only the material held in each primary packing/ink bag (14,16,18,20/32).

In operation, the nozzle array 42 can deposit materials from each primary packing/ink bag (14,16,18,20/32) individually, or multiple nozzles of an array 42 can operate to blend materials together, e.g. colours of inks or paints, to deposit at the same time. The bags 32 may contain different colours of marking materials, i.e. inks or paints, which may comprise CYM or, if good black is required, CYMK colours. Since the substrate or ground to have these deposited upon will not likely be white, a white may be required for any print that has white or a paler shade than the colours contained in the bags 32. When depositing ink or paint to print an image, the image may be printed in sweeps to generate small adjacent dots (i.e. each dot comes from a single nozzle of the array 42), and when viewed from above or a suitable distance from afar (e.g. from the stand in a stadium or from a television view) appearto blend into colours, depending on the relative colours of the different inks or colours deposited.

The flexible ink bag 32 comprises the red ink R suitable for depositing a red colour on a ground yet the flexible ink bag 32 may comprise any material for deposition, for example a marking material or a chemical to deposit on the ground, such as a herbicide, pesticide, insecticide, paint, ink, coloured material, powder, fertilizer, plant growth aid or water, or the like provided that a compatible hose 36 and nozzle arrays 42 are attached.

When printing an image, the flexible ink bags 32 will usually contain sufficient material to be able to print an entire image without changing during the printing run of the autonomous ground marking robot 10. However, if required, an ink bag can be changed during the deposition cycle.

The hose 36 is connected to a manifold 44 connected to a tank 46 containing chemical liquids 48 which serve a variety of purposes. The chemical liquids 48 may be used to flush the hose 36 and nozzles 42, increase or decrease the viscosity of the ink R or ground marking material by suitable mixing and may add effects to standard inks such as luminescent properties or change the chemical make-up of the ink or ground marking material.

In operation, a user receives a package containing primary packaging 14 as a lightweight, substantially rigid cardboard box containing therein a flexible bag 32 filled with a ground marking material, for example a red ink R. The user may register the marking material using the ID tag 30 to match marking materials held in a database by way of communication with module 22.

The database may contain a list of verified marking materials authorised for use and may in return grant permission for the autonomous ground marking robot 10 to accept the material and may, depending in the type of material, make mechanical or software adjustments. For example, a nozzle 42 height may be adjusted to spray fertilizer in a different way to that nozzle 42 arrangement for high resolution image printing. The database may also comprise a revocation list of packaging or materials that are no longer supported, out of date or out of contract. In which case, an error message may be displayed to the user.

In embodiments, the user may insert the primary packaging 14 into a frame 26 of the autonomous ground marking robot 10. Alternatively, if the arrangement allows, then the user may remove the ink bag 32 from the primary packaging 14 and place the ink bag into the frame 26. The sensor 28 may register the presence of the primary packaging 14 and further verify that the correct ink bag 32 is located in the correct frame and may further undertake a verified check of the authenticity of the ink bag 32 using RFID technology or measurement from the weight monitoring plate 14a. During printing or ground marking the weight of the ink bag 32 will decrease as ink is deposited onto the ground. The weight monitoring plate 14a can measure the change in weight and gather data.

The hose 36 is attached to the valve 34 and with appropriate setting up of the autonomous ground marking robot 10 as best described in Figures 3a, 3b, 4 and 5, printing or marking can commence.

Figures 3a and 3b are plan views of the autonomous ground marking robot 10, 3a is a top view, 3b is an underneath view, Figure 4 is a side elevation and Figure 5 is a plan view of a ground marking operation in progress, in this embodiment the tiled printing of an image.

The autonomous ground marking robot 10 comprises the case 12 held securely by a chassis supporting the ground wheel arrangement 24 with a print head 60 on a traverse guide 62, the traverse guide 62 permitting movement of the print head 60 beyond the width W of the ground wheel arrangement 24, along the length of the print width 68. The nozzle array 42 as described above may be attached to the print head 60. The nozzles maybe fixed and the print head 60 moveable. The print head 60, via the print guide 62, may be moveable along the length of a print width 68, which is the area the print head is capable of printing. The print head 60 many also be movable vertically based on the image to be printed, for example the print head 60 can be moved up and down depending on the density of the image to be printed.

The autonomous ground marking robot 10 further comprises a fisheye camera 13, wherein the camera 13 mounted inside of the machine case 12, on the top right side aligning with the first printing nozzle of the nozzle array 42. The camera 13 can be used to monitor the ground height and adjust the printhead height accordingly, allowing for more accurate image printing or material deposition. The camera 13 may also be used for tracking the progress of the deposition, or print if a technique, such as tiling, is required to print an image larger than the width of the autonomous ground marking robot 10, as shall be described with reference to Figures 7 and 8. Other suitable sensor technologies may also be used.

The ground wheel arrangement 24 comprises wheels 24a, 24b, 24c and 24d to steer the autonomous ground marking robot 10 along a path to affect the printing, and this may be under the control of a print file that can be loaded into an on-board control system such as may be contained in a smart communications module 22 (as will be described further with reference to Figures 9 & 10). The traverse guide 62 is fixed in relation to the ground wheel arrangement 24, so that it prints one line of an image along the print width 68. The ground wheel arrangement 24 then notches forward, moving the whole autonomous ground marking robot 10 forward for it to then print another line.

In another arrangement, also not illustrated, the traverse guide 62 can be movable relative to the ground wheel arrangement 24 in the direction of travel, so that an area may be printed while the ground wheel arrangement 24 is stationary, and then the ground wheel arrangement 24 moves forward by the length of the area printed, to then print an adjacent area of image. The print head 60 can, for example, print a line of 10mm width, then the ground wheel arrangement 24 notch forward by 10mm. Or an area, say of A4 or A3 paper size can be printed and only then does the autonomous ground marking robot 10 move forward.

The autonomous ground marking robot 10 can also print a strip 64, as shown in Figure 3, of image wider than the width W of the ground wheel arrangement 24 and when an entire strip 64 of image has been printed turn around to print an adjacent strip, in the art this is known as 'tiling'. In this way, the ground wheel arrangement 24 does not run over any part of the freshly painted ground, the outer tracks 66 of the ground wheel arrangement 24 being seen in Figure 3 to be well within the width of the strips 64.

The wheel arrangement 24 may have independent drives to manage torque for optimised positioning accuracy on any surface. The independent drives may be connected to the smart communications module 22 in order to feedback into drive control. The autonomous ground marking robot 10 may be able to respond in real time to changing terrain needs. The autonomous ground marking robot 10 may include an autonomous traction management capability, to safeguard the terrain the robot is interacting with and to reduce skidding and turf damage.

The print head 60 can be height adjustable, whereby to print finer or coarser images or to adapt to ground irregularities. The print head 60 can use any of a variety of printing techniques, including standard ink jet, spray, and 3D printing techniques involving melting plastic and dropping or shooting it at a ground surface.

All of these techniques are needed in order to satisfy a need in large image prints of an accuracy of tiling alignment should be +/-2mm and scalable to 50m² image/print activations. Thus, there is the need to control the tiling error to within ±2mm.

Figure 6 is a diagram of a simple image which is difficult to print in a tiled manner. There is shown an image 200 of a smiley face and grid 250 showing where the tiled sections could be for a very large version of the image 200. As can be seen, the image 200 has no flat edges and known techniques of following the print edge of a previous section cannot be used. Also for complex image prints where a print has section of zero marking, for example where there are gaps of paint in a print, it can be difficult to follow the edge of an existing print.

It is also very difficult in a tiled print using global positioning or beacon navigational techniques to line the eyes up on the smiley face. If the eyes were to be out of line across the print, even by a small amount, the human eye can pick up that they are not aligned. Also, if any of the round edges of the image are not matching to the previous edge of a previous tile, then this can be seen also by a human, especially if printed in a large size. Thus, a technique of being able to follow a 'true path' which is created as each section of a deposition or print takes place, needs to be used, as described with reference to Figures 7 & 8.

In Figure 7, there is shown a diagram of a sequence of dots with a specific pattern and an example camera view, according to one embodiment of the present invention. A plan view of an autonomous vehicle with a computer vision system, the field of view 13 of the computer vision system is shown as covering a certain number of dots and gaps, in this example a specific pattern 20 comprising 4 dots and a gap is shown.

As the autonomous machine prints a row during a first section, it lays down a series of dots and gaps in a repeating pattern 20. The machine prints a dot if the row number is divisible by 5, except when the row number is divisible by 25, when it doesn't print anything. Thus, a dot is printed every 5 rows, but not on the 25^th row, thus a pattern of 4 dots and gap is produced. This specific pattern is used so the whole pattern can be seen within the field of view 13 of the camera, such that the machine always knows whether it is in the middle, the start or end of the pattern.

Wherein the computer vision system calculates the machine position within a current print row, based on the dots printed from a previous section in a specific pattern 20 on the floor, comparing with the camera position (shown as a field of view 13) to determine how much should the machine moving forward. Thus, the machine prints dots with specific pattern 20 in the previous tile, the computer vision system locates the machine's real position after turning, based on the dots and other navigational information about its expected position.

As the machine knows that is has just printed the current row, the distance between each row from first tile and its camera position, it can calculate where it is within a deposition section and thus adjust its position should there be an offset from that true position. As shall be described in further detail in relation to Figure 8 following.

It should be clear to someone skilled in the art, that any colour paint can be used to deposit the dots as used in the image, so the dots are hidden by the next row of the image. If this is not possible due to a complex or fine resolution print, a UV paint can be used. Alternatively, deposited paint can be cleaned off by hand using a paint removal solution after a print.

It should also be clear to someone skilled in the art that and dots or dashes or other code or pattern can be used to allow the paint deposition machine to follow its own path in an accurate way. For example, a sticky QR code could be deposited by a labelling machine fixed to the side of the printer, for example if printing lines in a large logistics warehouse and the codes are required later by other machinery. Numbers or letters could be printed, other shapes such as squares or stars. The size of the dots could also get larger and then smaller again. Any code or pattern can be used in order to discern the machine's 'true path' in relation to the path it has already taken, such that an accurate sequential section can be deposited.

Figures 8 is a process flow diagram of the pattern following solution, according to one embodiment of the present invention. The steps of the method shall herein be described in further detail.

Process Start

SI: Take an image of the ground. As a line of a print is being deposited on the ground, the camera 13 of Figures 3-5 is able to take an image of the ground as the printer moves along. Each frame of the camera image is captured and processed by an on-board computer (with reference to Figures 9 & 10) in step 2.

S2: Process the image. Each frame of the camera image that is captured in Step 1 is processed by the on-board computer, as described in the following steps.

S3: Segment the dots and find the centrums (centres), dot sizes and gaps. This is so subsequent calculations are centralised to the centre of the dot and are not offset by the size of the dots themselves. The dot sizes and centrums are then used to find out if any adjustment is needed in the nozzle print height for the next print row also (see next step). The gap positions are used to determine the overall place in the pattern is at the beginning, middle or end.

S4: Analyse the dots and gaps pattern. This is done in order to calculate the current row position and based on the dots printed from a previous section in a specific pattern on the floor, comparing with the camera position to determine how much should the machine moving forward.

S5 Determine any x, y or z adjustments required. As the machine knows that is has just printed the current row, the distance between each row from first tile and its camera position (Step 4), it can calculate where it is within a deposition section and thus determine any offsets from that true position. Due to the calculation made of the centrums, gaps and sizes, the machine can also detect whether any height adjustments are necessary, due to a calculation of the size of the dots in the pattern and whether they are getting smaller of larger, for example.

S6: Re-adjust position if needed. Move forward by the required amount and in the required orientation. Making any adjustments to the print head height accordingly.

S7: Print the next row of the image. Plus, also deposit any dots or patterns required for the subsequent section. In the example, as described with reference to Figure 7, the machine prints a dot if the row number is divisible by 5, except when the row number is divisible by 25, when it doesn't print anything.

S8: Repeat process steps 1-7 for each row in the print until that tile segment is completed. Turn and repeat for the next tiled segment (if necessary).

End Process

Thus, the computer vision system locates the machine's real position after turning, based on the dots and other navigational information about its expected position.

Turning to Figure 9, a smart communications module 22 includes processing circuitry 80 coupled to memory circuitry 82, for example volatile memory (V)/non-volatile memory (NV), such as such as flash and ROM.

The memory circuitry 82 may store programs executed by the processing circuitry 80, as well as data such as user interface resources, time-series data, credentials (for example cryptographic keys) and/or identifiers for the remote resource (which may for convenience be referred to as the cloud 100 or the edge 102(s) (for example a URL or IP address). The memory circuitry 80 may also comprise access to machine learning algorithms stored in libraries to provide for an artificial intelligence equipped autonomous ground marking robot 10. The module 22 may also comprise input/output (I/O) circuitry 88 such as sensing circuitry to sense inputs (e.g. via sensors (not shown)) from the surrounding environment and/or to provide an output to a user e.g. using a buzzer or light emitting diode(s) (not shown). The module 22 may generate operational data based on the sensed inputs, whereby the operational data may be stored in memory 82. The I/O circuitry 88 may also comprise a user interface e.g. buttons (not shown) to allow the user to interact with the module 22.

The processing circuitry 80 may control various processing operations performed by the module 22 e.g. encryption of data, communication, processing of applications stored in the memory circuitry 82.

The module 22 may generate operational data based on the sensed inputs. For example, the module 22, may, for example, be an embedded device such as an ink registration and ink consumption monitoring device, which generates operational data related to the registration of an input primary packaging 14 comprising an ink bag 32 and the use of the ink R using data generated from the sensor 28 connected to a change in weight detected by the weight monitoring plate 14a.

Alternatively, the module 22 may, for example, comprise an embedded temperature sensor, which generates operational data based on the temperature of the surrounding environment, and may, for example be generated as a time series and fed, as best seen in Figure 10, to a remote resource such as the cloud 100, the edge 102 such as a tablet used to control the robot 10 via communications link 86.

The cloud 100 or the edge 102 may by return send instructions back to the autonomous ground marking robot 10 via a communications link 104 for the real-time adjustment of ground marking properties based on the data. In the present example, the cloud 100 and edge 102 may also communicate with each other via a communications link 108 and 110. This could be to update the instructions, send new instructions, initiate, or prevent the operation of the autonomous ground marking robot 10. The edge 202 may be between the communication between the autonomous ground marking robot 10 and the cloud 102. The robot laser 40 and position sensor 38 may communicate with the cloud 100 and/or the edge 102 to feedback into the real-time adjustment of ground marking properties based on the data.

Alternatively, the module 22 may, for example, comprise an accelerometer which generates data relating to the movement of the autonomous ground marking robot 10, for example capturing distance moved, or elevation ascended/descended by the autonomous ground marking robot 10 and fed to the cloud 100 or edge 102 for analysis.

Figure 10 schematically shows an example of the autonomous ground marking robot 10 in communication with the cloud 100, the edge 102, such as remote resource, which may be a tablet, smartphone or laptop when the present techniques are applied. The edge 102 may be a tablet controlled by a user, such as a Groundsman located on site responsible for the upkeep of a pitch within a football or rugby stadium.

In the present example, it will be appreciated that the cloud 100 may comprise any suitable data processing device or embedded system which can be accessed from another platform such as a remote computer, content aggregator or cloud platform which receives data posted by the autonomous ground marking robot 10. Use of a cloud 100 means that the onboard memory 82 of the robot does not need to store everything, any operational data for example any machine learning libraries, print instructions and operation instructions, and/or historical logging data, can be stored in the cloud 100.

In the present example, the autonomous ground marking robot 10 is configured to connect with the cloud 100 or the edge 102 to push data thereto, whereby, for the example, the autonomous ground marking robot 10 may be provided with the connectivity data (for example a location identifier (for example an address URL)) and credential data of the cloud 100 or the edge 102.

In the present example, on powering on for the first time, the autonomous ground marking robot 10 undertakes a registration process with the cloud 100 and the edge 102 and pushes identification data and is on standby to receive printing or ground marking data in return. It will be appreciated that the autonomous ground marking robot 10 may connect to the cloud 100 or the edge 102, e.g. via the internet, using one or more nodes/routers in a network e.g. a mesh network.

In alternative embodiments, a user may specify to which remote resource the autonomous ground marking robot 10 should push data. For example, the user may connect the autonomous ground marking robot 10 directly to a portable device e.g. via universal serial bus (USB), and install code capable of executing on the autonomous ground marking robot 10, whereby the code may comprise connectivity data and/or credential data relating to the remote resource with which the user wants the robot to communicate. The connectivity data and/or credential data may be provided to the autonomous ground marking robot 10 using any suitable method e.g. via USB/BLE. The credential data may also comprise credential data relating to a network to which the autonomous ground marking robot 10 may be required to connect, for example a WPA2 key for pairing with nodes in a WiFi network.

It will be clear to one skilled in the art that many improvements and modifications can be made to the foregoing exemplary embodiments without departing from the scope of the present technique. The robots, systems, and methods described herein can be adapted for use with different types of surface of substrate, depending on the purpose and surface for it to be used with.

The robots, systems, and methods described herein can be used to deposit material on multiple different substrates, surfaces, or the ground. For example, these could be, grass, turf, AstroTurf, artificial turf, synthetic turf, plastic turf, concrete, polished concrete, tarmac or tarmacadam ground surfaces, dirt, gravel, wood chip, carpeting, rubber, roads, asphalt, brick, sand, beaches, mud, clay wood, decking, tiling, stone, rock and rock formations of varying types of rock or stone, snow, ice, ice rinks, artificial snow, polymer surfaces such as polyurethane, plastic, glass and leather. The robots, systems, and methods described herein can be adapted for use with different surfaces, such as sports (e.g. football, cricket, racing, rugby, hockey, ice hockey, skiing, shooting) pitches, ski slopes, dry ski slopes, race courses, gymnasiums, indoor sports venues and running tracks.

In further exemplary embodiments, the robots, systems, and methods described herein may be used for printing or painting on a substrate or on the ground. This can be to print or paint, with inks or paint, logos, information, advertising or messages on the ground. When large images are printed, they are printed with adjacent dots or pixels so that when viewed from above or a suitable distance from afar (e.g. from the stand in a stadium or from a television view) the images are easily determined. Print instructions can be determined so that when an image, e.g. a logo is printed, they can be visible from stadium stand or by a viewer watching an event at home on television. The robots, systems, and methods described herein offer an improvement to printing methods for advertising purposes. Brand logos, slogans, pictures etc. can be printed to advertise a brand logo, image or message. These can be printed more efficiently, quickly and with a higher degree of accuracy than the methods and printers of the prior art.

The robot is therefore in some embodiments configured to print an image or logo on a surface, the robot housing two, three, four or more flexible bags containing a material for deposition, the material for deposition contained within each flexible bag being an ink or paint selected from a cyan, magenta, yellow, black, white, green, blue or red colour, the image or logo optionally being an advertising logo, design or safety warning. In further exemplary embodiments, the robots, systems, and methods described herein may be used for deposition of a fertiliser, pesticide or other such chemical for treatment of a substrate such as grass. The robots and method of using such robots described herein may also carry out multiple functions at the same time. For example, bags may contain paint for deposition to mark a logo on a pitch and may also contain fertiliser to fertilise the pitch.

Those skilled in the art will appreciate that while the foregoing has described what is considered to be the best mode and where appropriate other modes of performing present techniques, the present techniques should not be limited to the specific configurations and methods disclosed in this description of the preferred embodiment. Those skilled in the art will recognise that present techniques have a broad range of applications, and that the embodiments may take a wide range of modifications without departing from any inventive concept as defined in the appended claims.

Claims

Claims:

1. An autonomous deposition apparatus, the autonomous deposition apparatus comprising: a. at least one receptacle to hold a deposition material; b. at least one deposition arrangement; c. a locomotion arrangement; d. a sensor arrangement; e. a control unit, the control unit operable to receive at least one deposition instruction; wherein the deposition instructions include instructions for a multi section deposition; f. the control unit controlling the least one deposition arrangement and the locomotion arrangement to autonomously deposit the deposition material onto a surface in a first section, wherein during that first section, a true path is also deposited by the autonomous deposition apparatus, and g. wherein in a second section of the deposition, the control unit uses the sensor arrangement to dynamically monitor the true path and correct the position of the autonomous vehicle in order to deposit the second section in line with the true path.

2. Apparatus according to any preceding claim, wherein the sensor system comprises a computer vision system.

3. Apparatus according to claim 2, wherein the computer vision system comprises a camera.

4. Apparatus according to any preceding claim, wherein the true path is a code.

5. Apparatus according to claim 4, wherein the code is a pattern.

6. Apparatus according to claim 4 or 5, wherein the code is a QR code.

7. Apparatus according to claim, wherein the pattern is a repeating pattern.

8. Apparatus according to claim 7, wherein the pattern is a repeating pattern of dots and spaces.

9. Apparatus according to claim 7 or 8, wherein the pattern can be seen in its entirety within the field of view of the computer vision system.

10. Apparatus according to claim 4 or 6, wherein the code can be seen in its entirety within the field of view of the computer vision system.

11. Apparatus according to any preceding claim, comprising dynamically determining an offset with regard to the true path and operating one or more locomotion arrangements to minimize said offset.

12. Apparatus according to claim 11, further comprising dynamically determining an angular offset of the rotation angle (b) between the true path and the apparatus and the control unit controlling one or more locomotion arrangements to minimize said angular offset.

13. Apparatus as claimed in any preceding claim, wherein the onboard control system is configured to transmit data to a remote resource, such as a cloud server, or an edge device optionally a tablet or smartphone.

14. Apparatus as claimed in any preceding claim, wherein the locomotion arrangement is a ground wheel arrangement.

15. Apparatus as claimed in any preceding claim further comprising a chassis with a nozzle array on a traverse guide, the traverse guide permitting movement of the nozzle array beyond the width of the ground wheel arrangement.

16. Apparatus according to claim 15, in which the traverse guide is fixed in relation to the ground wheel arrangement.

17. Apparatus according to claim 16, in which the traverse guide is movable relative to the ground wheel arrangement in the direction of travel, so that an area can be printed while the ground wheel arrangement is stationary.

18. Apparatus as claimed in any preceding claim, wherein the material for deposition is a herbicide, pesticide, insecticide, plant growth aid, water or marking material, optionally wherein the marking material is a paint, ink, coloured material, powder.

19. Apparatus as claimed in any preceding claim, wherein the autonomous deposition apparatus is configured with a nozzle array to print an image or logo on a surface, the robot housing two or more flexible bags containing a material for deposition, the material for deposition contained within each flexible bag being an ink or paint selected from a cyan, magenta, yellow, black, white green, blue or red colour, the image or logo optionally being an advertising logo, design or safety warning.

20. A method of depositing material using apparatus of any one of claims 1 to 19, comprising a. a user sending deposition instructions to the autonomous deposition apparatus; b. and c. the autonomous deposition apparatus depositing material according to the deposition instructions.

21. A method as claimed in claim 20, wherein the deposition instructions are a command to print an image in a certain size and the control unit calculates the required sections of the print.

22. A method as claimed in claim 20 or 21, wherein the user sends deposition instructions to the autonomous deposition apparatus via a cloud server or device, or an edge server or device.