Automated concrete removal method and system

Abstract

Disclosed is a computer-implemented method for concrete removal using a robotic system (102) equipped with a high-pressure water nozzle (110). The method comprises receiving, via a sensor system (120), first image data including an area of interest in the vicinity of the robotic system (102), determining, based on the first image data, a concrete surface to be removed within the area of interest, defining (208) a coverage path for the high-pressure water nozzle (110) to traverse the concrete surface to be removed, wherein said defining includes determining waypoints and / or corresponding nozzle orientations to achieve an optimized path that covers the concrete surface to be removed, and causing executing (214) the concrete removal by the robotic system (102) along the coverage path, wherein real-time distance sensing is utilized to essentially maintain a predefined nozzle-to-surface distance, preferably in order to achieve a predetermined removal depth.

Description

TECHNICAL FIELD

[0001] The present invention relates to the field of autonomous robots that operate in unstructured environments, in particular in the construction industry to perform tasks such as hydrodemolition and concrete sanitation.BACKGROUND

[0002] In the field of autonomous robots operating in unstructured environments, such as for construction or concrete sanitation, companies typically offer teleoperation and semi-automated processing capabilities designed for planar concrete surfaces. The robots are typically based on tracked vehicles that move backwards once the front end-effector has traversed the assigned distance from one side to the other. To traverse a wanted surface, a worker must manually steer the machine to the desired location and orientation before starting the process.

[0003] While these solutions have shown promise in certain applications, they are often limited by their size and inflexibility. As a result, they struggle to effectively remove concrete on complex surfaces such as columns or corners, requiring manual intervention to complete the task. This not only increases labor costs but also poses safety and health risks for workers.

[0004] In contrast, other systems like the Woma Jetty offer more advanced capabilities, including six degrees of freedom (DOF) that enable greater flexibility and maneuverability. However, these solutions still rely on teleoperation, which can be challenging due to the complexity of controlling such a system manually.

[0005] US 2001 / 011841 A1 titled "Multiple jet hydrodemolition apparatus and method" discloses an apparatus for the hydrodemolition of concrete layer including a movable vehicle, a bed having a guideway extending transversely to a direction of movement of the vehicle, a nozzle assembly having a guide slidably engaging the guideway and a plurality of nozzles spaced apart in a direction transverse to the guideway, separate fluid flow controllers coupled between a pressurized source of fluid and respective nozzles, the nozzles being oriented to spray a fluid jet onto the concrete surface and means for moving the nozzle assembly along the guideway transversely across the direction of travel.

[0006] CA2879311A1 titled "SHIELDING SYSTEM FOR HYDRODEMOLITION APPARATUS" discloses a shield system for a hydrodemolition apparatus which has a loop whose ends are secured to opposite sides of a reciprocating nozzle carrier. The loop may be supported by a supporting structure surrounding the frame along which the nozzle carrier reciprocates. The loop overlays portions of one or more gaps through which one or more nozzles extend to work an underlying surface. The nozzles have access to the underlying surface through the gaps under the nozzle carrier, but the balance of the gaps are covered by the loop which is drawn or pushed by the nozzle carrier during reciprocation.

[0007] CN115726547A titled "Control method and device for automatically spraying wall body" discloses a control method and device for automatically spraying a wall. The method comprises obtaining a target wall image, recognizing a to-be-sprayed region in the target wall image, and decomposing the to-be-sprayed region into a plurality of sub-regions. A full-coverage spraying path in each sub-region is calculated, an adjacent graph is constructed according to the multiple sub-regions, and each sub-region is a node in the adjacent graph. The method further comprises performing path planning on each sub-region according to the adjacent graph to obtain an optimal path connecting an end point of each sub-region and a starting point of a next sub-region. The to-be-sprayed area is sprayed according to the optimal path and the full-coverage spraying path.

[0008] The existing state-of-the-art is characterized by several shortcomings. Firstly, current autonomous robots are unable to reliably perceive their environment or plan their own path, relying instead largely on manual intervention for navigation and control. This limits their ability to adapt to changing conditions or unexpected obstacles. Secondly, the machines' limited degrees of freedom make them difficult to use and maneuver in tight spaces or complex environments. This inflexibility can lead to reduced productivity and increased risk of damage or injury. Thirdly, many current solutions are not well-suited for processing hard-to-reach surfaces, requiring manual intervention or specialized equipment to access these areas. Fourthly, the machines' size and rigidity often make them difficult to fit into tight spaces, further limiting their effectiveness in certain applications. Lastly, most autonomous robots in this field require teleoperation by a worker, which can be time-consuming and labor-intensive. This reliance on human intervention limits the machines' autonomy and effectiveness.

[0009] In summary, while current solutions have shown promise in certain applications, they are often limited by their size, inflexibility, and lack of adaptability to complex surfaces or environments. The need for more advanced autonomous robots that can effectively process unstructured environments remains a pressing challenge in industries such as construction and concrete sanitation.

[0010] It is therefore a technical problem underlying the present invention to provide improved techniques for concrete deconstruction which are in particular more efficient.SUMMARY OF THE INVENTION

[0011] The problem is solved by the subject-matter defined in the independent claims. Advantageous modifications of embodiments of the invention are defined in the dependent claims as well as in the description and the figures.

[0012] According to a first aspect of the present invention, a method for concrete removal is provided. The method may be computer-implemented. The concrete removal may be performed using a robotic system, also referred to herein simply as robot or as mobile manipulator. The robotic system may be equipped with a high-pressure water nozzle.

[0013] The method may comprise a step a) of receiving, via a sensor system, first image data including an area of interest in the vicinity of the robotic system. The method may comprise a step b) of determining, based on the first image data, a concrete surface to be removed within the area of interest. The method may comprise a step c) of defining a coverage path for the high-pressure water nozzle to traverse the concrete surface to be removed. Said defining may include determining waypoints and / or corresponding nozzle orientations to achieve an optimized path that covers the concrete surface to be removed. The method may comprise a step d) of causing executing the concrete removal by the robotic system along the coverage path. Real-time distance sensing may be utilized to essentially maintain a predefined nozzle-to-surface distance, preferably in order to achieve a predetermined removal depth.

[0014] The term "robotic system" may be understood as an automated or semi-automated machine with one or more articulated arms, mobility platforms, or any combination thereof, specifically designed at least for concrete removal. The robotic system may include various sensors and actuator components configured to execute programmed tasks with precision.

[0015] The term "high-pressure water nozzle" may be understood as a nozzle capable of directing a jet of water at high pressures, which may be utilized as a means for concrete removal. The nozzle may be attached to the robotic system and may be controlled to move along a predefined path for effectively breaking up and removing concrete.

[0016] The term "sensor system" may be understood as at least one sensor, in particular a collection of sensors, such as cameras, LiDAR, ultrasonic sensors, and / or others, integrated in and / or being communicatively coupled to the robotic system. The sensor system may be responsible for capturing first image data and may also provide real-time distance measurements during the removal operation.

[0017] The term "image data" may be understood as a visual representation obtained from the sensor system that includes an area of interest in the vicinity of the robotic system. The data can be in the form of digital images and / or three-dimensional point clouds. It may be used to identify the concrete surface requiring removal.

[0018] The term "area of interest" may be understood as a specific zone within the vicinity of the robotic system where the robotic system is physically located. The area of interest includes a concrete surface that is targeted to be removed by executing the concrete removal process.

[0019] The term "concrete surface to be removed" may be understood as a particular section of concrete identified within the area of interest that has been determined to require removal.

[0020] The term "coverage path" may be understood as a strategically calculated route that the high-pressure water nozzle follows during the concrete removal process. The path ensures that the entire surface designated for removal is covered effectively. The term "waypoints" may include a set of spatial coordinates that define critical points along the coverage path which the high-pressure water nozzle follows.

[0021] The term "nozzle orientation" may be understood as an angular position of the high-pressure water nozzle relative to the concrete surface in three-dimensional space intended to maximize the efficiency of concrete removal. The nozzle orientation may be controlled by an arm of the robot system and / or by one or more hinge elements and / or the like.

[0022] The term "optimized path" may be understood as a particularly efficient route that has been calculated for the nozzle to traverse in order to remove essentially the entire predefined concrete surface. This may involve minimizing the time and / or water usage and / or energy required for the task, while achieving the desired level of concrete removal.

[0023] The term "executing concrete removal" indicates the operational phase where the robotic system carries out the actual physical process of removing the concrete using the high-pressure water nozzle, following the defined coverage path.

[0024] The term "real-time distance sensing" may be understood as an ongoing process of measuring the distance between the high-pressure water nozzle and the concrete surface during the removal operation, to ensure consistent removal depth and enhance overall precision.

[0025] The term "predefined nozzle-to-surface distance" may be understood as a distance that may be maintained between the nozzle and concrete surface to adequately achieve a predetermined level of removal depth without causing unnecessary damage to the underlying structure.

[0026] The term "predetermined removal depth" may be understood as a target depth to which the concrete needs to be removed, which can vary depending on the specific requirements of the application or construction project.

[0027] The combination of advanced imaging and sensor systems to receive first image data that includes the area of interest allows for precise identification and determination of the concrete surface needing removal. Consequently, the method averts the indiscriminate destruction of material, reducing waste and preserving structural integrity where necessary.

[0028] Further, the step of defining an optimized coverage path for the high-pressure water nozzle to traverse the concrete surface targeted for removal which includes the determination of waypoints and corresponding nozzle orientations, not only ensures thorough coverage of the concrete surface for removal but also significantly reduces the time and resources required to complete the task. As a result, the method enhances efficiency and can contribute to reduced operational costs.

[0029] Executing the concrete removal along the defined coverage path, the inventive process incorporates real-time distance sensing to maintain a predefined nozzle-to-surface distance consistently. This feature facilitates achieving a predetermined removal depth, which is essential for tasks requiring a high precision, such as in the preservation of underlying structural elements or when preparing surfaces for subsequent treatments or finishes. By maintaining consistent pressure and distance, the method minimizes variability in removal depth and avoids structural weaknesses that might arise from irregular surface treatment.

[0030] The use of real-time distance sensing also serves to accommodate for variations in the working environment, such as uneven surfaces or accessibility challenges. By adjusting the robotic system's operation dynamically, the method mitigates the risk of incomplete or excessive removal that could result in project delays or the need for costly rework.

[0031] Thus, an autonomous and efficient method for concrete removal using a robotic system is provided. This creates several further technical advantages: Better extraction selection, because the robot can either scan the environment or detect on its own what surface to process. The interface of the existing machines is usually difficult to operate, and teleoperating an end-effector does not work with good quality. The method can provide better extraction quality, due to the robot being able to hold the power of the water thrust better than a human. When looking at existing solutions, there are either workers doing the task manually, standard systems not being able to remove complex shapes due to their design, and systems which are teleoperated with a similar concept as ours. Humans cannot hold the water lance as accurate as our system due to the thrust of the water. Most known systems are not designed to be flexible but instead only for large, planar concrete removal. These systems therefore cannot extract concrete from round columns or corners for example. Compared to systems such as the Woma Jetty, embodiments of the invention can also be more accurate because the robot of the present invention is actuated with electricity and not with hydraulics, which are difficult to control. Current hydraulic arms are simply controlled by teleoperation, making control difficult and inefficient. Furthermore, with real-time distance measurement and control of the water lance to the surface, the robot can better adapt to the processing area than any current machines or humans. Iteratively improving results, because the robot can scan the wall after it processed the complete area, to see how deep it reached. If there is not enough concrete removed, the robot can again on its own plan a path to remove the rest of the concrete which was not done by the first sweep. Currently, this all must be done by manual labor. Bigger flexibility in the environment, due to the smaller robot with an arm with at least six DoF. The robot will be able to fit through doorways and can also work in environments where there is a lot of clutter and tight passages. With collision avoidance, we will be able to also let the robot work where current machines do not fit. Faster process, because the robot does not need to take a rest compared to humans.

[0032] Also, it can handle higher thrusts than humans, so that it can process surfaces faster. With autonomous movement, the robot furthermore will be able to place itself into the optimal positions for removing concrete, so the worker does not have to move the robot after every done extraction. Safer for humans, because manual labor can now be automated by a robotic solution. When working with current machines, not every surface can be processed, so manual labor sometimes still is necessary. Can be transported without a big truck, making the handling of the robot much easier. As the system is light enough to put in a van, no lorry license is required from a worker to bring the machine to the construction site. This makes the operation of the robot much easier to organize than with current systems. The robot can work with a small pump instead of container pump. This means that logistics are easier when not having to organize a space for the container to sit in, as well as being able to use normal water access instead of needing a water hydrant.

[0033] It may be provided that the method further comprises a step e) of, after executing the concrete removal, receiving, via the sensor system, second image data including the area of interest in the vicinity of the robotic system to detect any portions of the concrete surface to be removed that require additional concrete removal, in particular based on the predetermined removal depth, wherein optionally a before-after comparison map is generated based on the first and second image data to evaluate the effectiveness of concrete removal. This way, real-time feedback and evaluation of the concrete removal process can be provided.

[0034] It may be provided that the method further comprises a step f) of iteratively re-planning and executing supplementary coverage paths for portions identified as requiring additional concrete removal until a uniform removal depth is achieved throughout the concrete surface to be removed. This way, an iterative refinement process is provided that ensures a uniform removal depth across the entire concrete surface. By iteratively re-planning and executing supplementary coverage paths for portions identified as requiring additional concrete removal, the robot can achieve precise control, because the robot's advanced planning capabilities allow it to generate optimized supplementary coverage paths that precisely target areas requiring additional removal. Furthermore, rework may be minimized, because iterative refinement reduces the need for manual intervention and minimizes rework, resulting in a more efficient cleaning process with reduced labor costs. Lastly, uniformity may be ensured because by continuously monitoring and adjusting its performance, the robot ensures a consistent removal depth throughout the concrete surface, eliminating hotspots or areas that may require additional attention.

[0035] It may be provided that the sensor system includes a 3D LiDAR sensor and a wide field-of-view RGB camera, preferably being integrated into the robotic system. This way, accurate spatial mapping, enhanced object detection and improved scene understanding may be achieved.

[0036] It may be provided that step b) of determining the concrete surface to be removed within the area of interest includes: (i) detecting, based on the image data, painted markings within the area of interest, the painted markings at least partially indicating the concrete surface to be removed, wherein the painted markings may be provided by a user using a painting device configured to mark concrete surfaces in real world, optionally wherein an electronic user interface device visualizes the area of interest including the painted markings; By detecting painted markings that indicate the concrete surface to be removed, the robot can more accurately determine what areas need to be cleaned. The detection of painted markings provides real-time feedback on the environment, allowing the robot to adjust its behavior and make informed decisions about where to focus cleaning efforts. By following the painted markings, the robot can optimize its cleaning path and reduce rework or manual intervention.

[0037] In addition or alternatively, It may be provided that step b) of determining the concrete surface to be removed within the area of interest includes: (ii) receiving user input via the electronic user interface device, wherein determining the concrete surface to be removed is performed based on the user input, wherein preferably, the user is provided with a visualization of the area of interest by the electronic user interface device based on which the user input can be performed. Allowing users to provide input via an electronic interface device enables them to customize the cleaning process based on specific requirements or preferences. User-provided input can help ensure that the robot accurately identifies the concrete surface to be removed, reducing errors and improving overall performance. Providing a visualization of the area of interest via an electronic interface device allows users to better understand the cleaning process and make informed decisions about where to focus their efforts. Therefore, in other words, areas of interest can be marked by a digital pen by tracing the circumference of the current area to process, or by marking surfaces on the electronic interface device which may be a control tablet.

[0038] It may be provided that the method further comprises visualizing, on the electronic user interface device, the defined coverage path for approval and / or refinement by the user. By providing a clear visualization of the defined coverage path, users can better understand how their input will be translated into action. Allowing users to review and refine the defined coverage path enables them to tailor the cleaning process to specific requirements or preferences. Visualizing the defined coverage path helps prevent mistakes by allowing users to identify potential issues before the robot begins cleaning. By refining the defined coverage path, users can optimize the cleaning process and reduce rework or manual intervention.

[0039] It may be provided that the method further comprises transitioning the robotic system to a safe state under certain safety conditions, received from a plurality of safety sensors, and as decided by a programmed safety system within the robotic system, optionally wherein the safe state includes stopping movement and / or stopping processing of the robotic system. The ability for the robot to automatically transition into a safe state in response to changing environmental or sensor data ensures that potential hazards are mitigated, reducing the risk of accidents or damage. By integrating multiple safety sensors and a programmed safety system, the robotic system can quickly respond to unexpected situations, such as obstacles or changes in its environment. The ability for the robot to transition into different safe states (e.g., stopping movement and / or processing) allows it to adapt to various scenarios, ensuring continued operation within defined safety parameters. By automating the process of transitioning to a safe state, the robotic system reduces the need for human intervention or manual override, simplifying overall control and reducing potential errors It may be provided that real-time distance sensing is performed using ultrasonic sensors. Ultrasonic sensors provide accurate measurements of distances, even in environments with varying lighting conditions or reflective surfaces. Real-time distance sensing enables the robotic system to receive continuous updates on its surroundings, allowing for more precise navigation and obstacle avoidance. Ultrasonic sensors can operate effectively over a wide range of distances (typically up to 10-15 meters), making them suitable for various applications. Ultrasonic sensors are less susceptible to electromagnetic interference or radio frequency interference, ensuring reliable distance sensing in environments with high levels of electrical activity.

[0040] It may be provided that method further comprises adjusting water pressure and / or flow rate based on the real-time distance sensing and / or the type of concrete surface encountered during the concrete removal. By adjusting water pressure and / or flow rate in response to changing distances or surface types, the robotic system can optimize its cleaning performance, reducing rework or manual intervention. Real-time adjustments enable the robot to tailor its cleaning approach to specific concrete surfaces, ensuring effective removal of dirt, grime, and other contaminants without damaging the surface. By optimizing water pressure and flow rate based on real-time sensing data, the robotic system can minimize water usage while still achieving effective cleaning results. The ability to adjust water pressure and / or flow rate in response to changing conditions enables the robot to adapt to various concrete surfaces, including those with different textures or porosities.

[0041] It may be provided that the execution according to step d) is performed autonomously by the robot system, preferably offline and / or without additional inputs from the user until completion of the coverage path or upon encountering a safety condition that triggers safe state transition. By automating this step, the robotic system can complete its cleaning task independently, reducing the need for human intervention and minimizing downtime. Autonomous operation reduces the risk of accidents or injuries caused by human error or distraction during the cleaning process. Offline autonomous operation enables the robot to adapt to changing environmental conditions or unexpected obstacles without requiring user input or manual override. By automating the step, the robotic system can reduce labor costs associated with manual intervention and minimize the need for human supervision.

[0042] It may be provided that the robotic system includes a robotic arm where the high-pressure water nozzle is located, the robotic arm having at least six degrees of freedom to manipulate the orientation of the high-pressure water nozzle during traversal of the coverage path. The ability to manipulate the orientation of the high-pressure water nozzle through multiple DOFs allows for more precise control and adaptability during traversal of the coverage path, enabling the robot to reach complex or hard-to-access areas. With a robotic arm capable of manipulating its orientation in six dimensions (e.g., x, y, z, roll, pitch, yaw), the system can achieve better coverage of the concrete surface by adjusting the nozzle's angle and position to ensure thorough cleaning. The increased flexibility provided by the robotic arm enables it to reach areas that would be difficult or impossible for a fixed-nozzle robot to access, such as corners, edges, or recessed surfaces. By precisely controlling the orientation of the high-pressure water nozzle, the system can minimize scratches and damage on the concrete surface, ensuring a smoother finish and reducing the risk of defects. The ability to manipulate the nozzle's orientation allows for more precise control over the cleaning process, enabling the robot to avoid sensitive areas or obstacles while maintaining a safe distance from them.

[0043] Based on the received first image data, a first point cloud of the area of interest, preferably a first colored point cloud of the area of interest, may be generated. The same may be performed for the second image data. For example, steps c), d), e), f), (i) and / or (ii) may be performed on the basis of such point cloud.

[0044] According to another aspect of the invention, a robot system for concrete removal is provided. The robot system is configured to perform any of the methods described herein. It may be provided that the robot system comprises a mobile base equipped with a vehicle platform for navigating various terrain types, the vehicle platform preferably being a tracked vehicle platform. It may be provided that the robotic system comprises a robotic arm coupled to the mobile base. The robotic arm may provide at least six degrees of freedom for maneuvering an attached high-pressure water nozzle designed to perform removal of concrete surfaces, wherein preferably the high-pressure water nozzle is interchangeably attached to the robotic arm, allowing for the substitution of the nozzle with other end-effectors for performing additional functions beyond concrete removal. It may be provided that the robotic system comprises a sensor system configured to scan an area of interest in the vicinity of the robot system. It may be provided that the robotic system comprises a processing unit communicatively coupled with the robotic arm and the sensor system. It may be provided that the robotic system comprises an electronic user interface device or a communication interface configured to communicate with an electronic user interface device. Optionally, it may be provided that the robotic system comprises a safety system equipped with safety sensors including radars for detecting humans and / or sensors for staircase fall protection.

[0045] According to another aspect of the invention, a data processing apparatus is provided. The data processing apparatus is configured to perform any of the methods described herein.

[0046] According to another aspect of the invention, a computer program or a computer-readable medium having stored thereon a computer program is provided. The computer program comprises instructions which, when the program is executed by a computer, preferably by a data processing apparatus as described above, more preferably by a robotic system as described above, cause the robotic system to carry out any of the methods described herein

[0047] All features, technical implementation details and advantages described with respect to any one of the aspects of the present invention described herein are self-evidently mutatis mutandis applicable for any one of the other aspects of the present invention and vice versa.BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The disclosure may be better understood by reference to the following drawings: Fig. 1:A schematic block diagram of a robotic system according to embodiments of the present invention. Fig. 2:A flow chart of a process for operating the robotic system according to embodiments of the present invention. Fig. 3:A photograph of marked and deteriorated concrete which needs to be removed according to embodiments of the present invention. Fig. 4:Images illustrating a process of finding a marked area and determining a trajectory for the autonomous robot to follow according to embodiments of the present invention. Fig. 5:A pseudocode listing of an algorithm for detecting 3D waypoints based on an image according to embodiments of the present invention. Fig. 6:A visualization of how to find waypoints for coverage traversal of a shape according to embodiments of the present invention. Fig. 7:A pseudocode listing of an algorithm for finding the horizontal pixels needed for shifting a vertical trajectory by a given step size in 3D according to embodiments of the present invention. DETAILED DESCRIPTION

[0049] Fig. 1 shows a schematic block diagram of a robotic system 102, also referred to as mobile manipulator or simply robot herein, according to an exemplary embodiment of the invention. The mobile manipulator 102 comprises multiple building blocks. The mobile basis 114 enables the robot 102 to move in its environment. The mobile basis 114 is preferably a tracked vehicle enabling the traversal of various kinds of ground structure. On the mobile basis 114, the rest of the robot 102 can be mounted in a modular fashion, to enable the usage of different robotic arms, end-effectors, as well as utility trunks for the robot 102.

[0050] The safety system 118 can be provided on a programmable logic controller (PLC) configured to determine if the robot 102 needs to go to a safe state to not endanger its environment. The safe state may comprise simply stopping the robot 102 so that it does not move. The safety system 118 is fed by a sensor system 120 including one or more sensors configured to determine the current safety state.

[0051] The sensors 120 of the robot 102 may comprise one or more depth cameras 122, one or more LiDARs 122, one or more RGB cameras 124 or any combination thereof. In a preferred embodiment, the sensors 120 the robot 102 comprises a 3D LiDAR 122 combined with wide-FoV RGB cameras 124 to collect a colored point cloud. The point cloud can be used for determining what area the robot 102 should process. The sensory 120 may also comprise safety sensors such as radars capable of scanning for humans in the environment. These sensors can be used for determining if the robot is in a safe state or not. Furthermore, there may be staircase fall protection sensors, distance sensors for the end-effector, ultrasonic sensors, or any other combination of these sensors.

[0052] The 3D LiDAR as well as the end-effector may be mounted on the robot arm 112. This way, the robot 112 can act as a manipulator in the environment, where it is able to autonomously scan the surroundings and plan paths on its own to process surfaces. The manipulator will be exchangeable to enable the usage of different arms for different use cases.

[0053] The computer 116, also referred to as processing unit herein, of the robot 102 is configured to gather sensor inputs and to calculate according to the current state the path the robot 102 should take. This may involve especially how coverage path planning is done for surface traversal.

[0054] The high-pressure water nozzle 110 is the end-effector in the illustrated embodiment, capable of removing concrete by ejecting water onto the surface in front. This end-effector can also be exchanged, so that the robot 102 can take over other tasks such as painting or gripping things.

[0055] Outside of the mobile manipulator system 102 is a water pump 104 providing pressurized water to the robot 102. The water may be fed through a high-pressure capable water hose to the robot 102.

[0056] A worker 106 may control the robotic system 102, e.g., with a control tablet 108 or another electronic user interface device.

[0057] Fig. 2 shows a flow chart illustrating a process of operating the robotic system 102. In certain embodiments, there is a worker 106 controlling the system to have a human-in-the-loop. In step 202, the worker 106 first steers the robot 102 to the wall where there is concrete to be removed. This may include unloading the robot 102 from a truck it is transported in, connecting it to the water hose 104 and electricity if needed, and steering it into the garage, for example.

[0058] In step 204, the robot 102 scans the current environment and displays it on the tablet 108 the worker 106 holds. The environmental scanning process must be done carefully in order to not crash into the surroundings, as these are unknown beforehand.

[0059] In step 206, the worker 106 can then digitally define an area on the control tablet 108 to indicate the area on which the robot 102 should work on. This can be done by either pressing on a wall and fully selecting it, or by marking a specific area and only doing path planning for this. Also, automatic detection of construction spray can be done so that only the detected area needs to be selected by one touch on the display.

[0060] In step 208, the robot 102 plans and visualizes the path it calculated. This is important to show the supervising worker 106 what the robot 102 is going to do. The worker 106 is given the chance to check the path and start the execution in step 210. If the planned path is not correct, the selection procedure of the wall can be repeated, and another path can be planned in step 208. If the surface is traversed correctly, the worker 106 can approve the procedure and it gets sent to the robot 102 for execution in step 214.

[0061] The high-pressure water concrete removal is then done offline by the robot 102. This means that during the procedure, the path is not recalculated, only distance sensors may be used to determine if the end-effector is too close or too far away from the wall. This is important because the nozzle should not crash into the wall to prevent damage to the hardware, but being too far away deteriorates the extraction quality.

[0062] In the following a variation of a concrete removal process according to an exemplary embodiment of the invention is explained from the point of view of the worker 106. 1. Construction worker 106 takes tablet control 108 and drives robot 102 out of van to place it for preparation. 2. Worker 106 takes water hose and electricity cable and attaches it to robot 102 to make it ready for concrete removal. 3. Worker 106 takes tablet 108 and drives robot 102 into construction site to place it in front of a surface which needs concrete removal. 4. Worker 106 presses "Scan" on tablet 108, so that the robot 102 scans its environment for creating a map. 5. Worker 106 a) Selects an area on the tablet 108 which was detected by the robot 102. b) Takes a digital marking device and circles the area to process. c) Circles the area to process on the tablet 108. to indicate the robot 102 what surface it should plan a trajectory on. 6. The robot 102 takes the marked surface in the form of a point cloud and plans a trajectory to traverse the complete area. 7. The robot 102 displays the trajectory it planned on the tablet 108 so that the worker 106 can check it. 8. The worker 106 checks the trajectory for any errors and either accepts it or tells the robot 102 to replan. For replanning, go back to step 5. 9. If the trajectory is accepted, the robot 102 receives the start command and begins to increase water pressure and flow, to start concrete removal. The worker 106 is at a safe distance and monitors the process. 10. After the robot 102 finishes the traversal, it moves back to the home position and waits for the moisture in the air and the dust to settle. 11. As soon as the vision is clear again, the robot 102 rescans the processed area to produce a before-after comparison. 12. If there are parts of the area which have not been extracted deep enough, the robot 102 plans another trajectory over only these parts to reprocess the surface to ensure equal extraction depth. Go back to step 7. 13. If the concrete extraction has been successful in every part of the area, the worker 106 takes the robot control tablet 108 and moves the robot 102 to the next position for further concrete removal tasks.

[0063] In the following, an exemplary implementation of the above concepts in accordance with an exemplary embodiment of the invention is described. The autonomous concrete removal robot 102 of this embodiment is configured to receive a point cloud and RGB camera information as inputs. The robot 102 is configured to recognize a marked shape outlined by an expert 106 with spray paint. An example is shown in Fig. 3. The robot 102 is configured to automatically assign waypoints onto the boundaries of the area.

[0064] The used algorithm enables the robot 102 to first detect the marked figure on a wall in front of the camera. After finding the shape, waypoints are determined and the corresponding normal orientations are calculated. Between each vertical stride, the horizontal step size is determined according to a specified distance. After finding all waypoints, the robot 102 is supposed to traverse them. During the entire operation, the robot 102 must avoid collisions with the wall and ground.

[0065] To do this, the sensor data is processed. The image becomes an OpenCV image, and the point cloud gets converted to an array of three dimensions, where the coordinates in space get saved at the corresponding pixel values in the array. The remaining process works with the image data gained. In the illustrated embodiment, the marked area is indicated by red spray paint. First, a Boolean mask is created, indicating the red pixels with true and the rest with false. Two arrays are specified that contain the lower and upper bounds of the HSV values we determine as marked. As the red color can be found on both edges of the hue spectrum, this is done twice, once for the lower edge and once for the higher edge of the range. A mask is then created by first finding the lower and higher masks and adding them. The process is illustrated in Fig. 4.

[0066] To set waypoints on the boundary of the marked area, the contour of the image is calculated, as we need it to fill the mask, by using the OpenCV function findContours(), where we need to give the mask as an argument. This contour is then filled with the function fillPoly(). Now that the filled area is determined from the image taken with the camera, the algorithm can start trying to set the waypoints to traverse the figure. One method for setting waypoints is described in the algorithm shown in Fig. 5.

[0067] In certain embodiment, finding the waypoints is performed as follows: First, find the outer bounding box of the mask. Then start inside this area in the upper left corner and loop vertically through the pixels. When a pixel is inside the shape, but its neighbor pixel above or below is outside, it is on the boundary of the marked area. We append the boundary pixel to a list and continue our loop. This search continues until we arrive at the y-position of the lowest extremum of the shape. Here we take a horizontal step to the right, specified by a step size with another method we will explain next. We then set our y-position to the highest point of the bounding box and repeat the described process until we take enough horizontal steps to arrive at the right extremum. The process is visualized in Fig. 6.

[0068] It is essential to have the correct step size, as we want even coverage of the water jet on the wall. To take a fitting step size, we implemented the method horizontal_step() depicted in Fig. 7. It receives the step size and current pixel coordinates on the image from which we want to take the step. The goal is to take a pixel step to the right until we reach the given spatial step size. The loop behaves as follows: If the covered distance is still less than the given step size, we move one pixel to the right, increasing the pixel's horizontal position. At this point, we calculate the spatial distance between the current pixel position and the position to the left. This distance gets added to the covered distance every time we pass the loop. Important to note is that we should not calculate the Euclidean distance in 3D space as pixels on the same pixel row might have different heights in space. We are only interested in the horizontal distance traveled. We should ignore the difference in the height of the coordinates and only calculate the Euclidean distance in the x - y plane.

[0069] As soon as the waypoints are determined, a trajectory can be calculated. This can be done by standard measures such as Rapidly Exploring Random Trees (RRT). While calculating the trajectory every robot pose can also be assessed for collisions with itself and the environment. The result is a robot trajectory which follows the given waypoints in order, without colliding with the environment.

[0070] Now that the robot was able to autonomously find a trajectory for concrete removal by scanning the environment and filtering for the market shape, the execution can be started. This may simply be done by giving the robot the respective trajectory and instructing it to follow the path.

[0071] Although some aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the corresponding method, where a block or device corresponds to a method step or a feature of a method step. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus.

[0072] Embodiments of the invention may be implemented on a computer system. The computer system may be a local computer device (e.g. personal computer, laptop, tablet computer or mobile phone) with one or more processors and one or more storage devices or may be a distributed computer system (e.g. a cloud computing system with one or more processors and one or more storage devices distributed at various locations, for example, at a local client and / or one or more remote server farms and / or data centers). The computer system may comprise any circuit or combination of circuits. In one embodiment, the computer system may include one or more processors which can be of any type. As used herein, processor may mean any type of computational circuit, such as but not limited to a microprocessor, a microcontroller, a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a graphics processor, a digital signal processor (DSP), multiple core processor, a field programmable gate array (FPGA), or any other type of processor or processing circuit. Other types of circuits that may be included in the computer system may be a custom circuit, an application-specific integrated circuit (ASIC), or the like, such as, for example, one or more circuits (such as a communication circuit) for use in wireless devices like mobile telephones, tablet computers, laptop computers, two-way radios, and similar electronic systems. The computer system may include one or more storage devices, which may include one or more memory elements suitable to the particular application, such as a main memory in the form of random access memory (RAM), one or more hard drives, and / or one or more drives that handle removable media such as compact disks (CD), flash memory cards, digital video disk (DVD), and the like. The computer system may also include a display device, one or more speakers, and a keyboard and / or controller, which can include a mouse, trackball, touch screen, voice-recognition device, or any other device that permits a system user to input information into and receive information from the computer system.

[0073] Some or all of the method steps may be executed by (or using) a hardware apparatus, like for example, a processor, a microprocessor, a programmable computer or an electronic circuit. In some embodiments, some one or more of the most important method steps may be executed by such an apparatus.

[0074] Depending on certain implementation requirements, embodiments of the invention can be implemented in hardware or in software. The implementation can be performed using a non-transitory storage medium such as a digital storage medium, for example a floppy disc, a DVD, a Blu-Ray, a CD, a ROM, a PROM, and EPROM, an EEPROM or a FLASH memory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable.

[0075] Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.

[0076] Generally, embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer. The program code may, for example, be stored on a machine readable carrier.

[0077] Other embodiments comprise the computer program for performing one of the methods described herein, stored on a machine readable carrier.

[0078] In other words, an embodiment of the present invention is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

[0079] A further embodiment of the present invention is, therefore, a storage medium (or a data carrier, or a computer-readable medium) comprising, stored thereon, the computer program for performing one of the methods described herein when it is performed by a processor. The data carrier, the digital storage medium or the recorded medium are typically tangible and / or non-transitory. A further embodiment of the present invention is an apparatus as described herein comprising a processor and the storage medium.

[0080] A further embodiment of the invention is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein. The data stream or the sequence of signals may, for example, be configured to be transferred via a data communication connection, for example, via the internet.

[0081] A further embodiment comprises a processing means, for example, a computer or a programmable logic device, configured to, or adapted to, perform one of the methods described herein.

[0082] A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0083] A further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of the methods described herein to a receiver. The receiver may, for example, be a computer, a mobile device, a memory device or the like. The apparatus or system may, for example, comprise a file server for transferring the computer program to the receiver.

[0084] In some embodiments, a programmable logic device (for example, a field programmable gate array) may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

Examples

Embodiment Construction

[0049]Fig. 1 shows a schematic block diagram of a robotic system 102, also referred to as mobile manipulator or simply robot herein, according to an exemplary embodiment of the invention. The mobile manipulator 102 comprises multiple building blocks. The mobile basis 114 enables the robot 102 to move in its environment. The mobile basis 114 is preferably a tracked vehicle enabling the traversal of various kinds of ground structure. On the mobile basis 114, the rest of the robot 102 can be mounted in a modular fashion, to enable the usage of different robotic arms, end-effectors, as well as utility trunks for the robot 102.

[0050]The safety system 118 can be provided on a programmable logic controller (PLC) configured to determine if the robot 102 needs to go to a safe state to not endanger its environment. The safe state may comprise simply stopping the robot 102 so that it does not move. The safety system 118 is fed by a sensor system 120 including one or more sensors configured to ...

TECHNICAL FIELD

[0001] The present invention relates to the field of autonomous robots that operate in unstructured environments, in particular in the construction industry to perform tasks such as hydrodemolition and concrete sanitation.BACKGROUND

[0002] In the field of autonomous robots operating in unstructured environments, such as for construction or concrete sanitation, companies typically offer teleoperation and semi-automated processing capabilities designed for planar concrete surfaces. The robots are typically based on tracked vehicles that move backwards once the front end-effector has traversed the assigned distance from one side to the other. To traverse a wanted surface, a worker must manually steer the machine to the desired location and orientation before starting the process.

[0003] While these solutions have shown promise in certain applications, they are often limited by their size and inflexibility. As a result, they struggle to effectively remove concrete on complex surfaces such as columns or corners, requiring manual intervention to complete the task. This not only increases labor costs but also poses safety and health risks for workers.

[0004] In contrast, other systems like the Woma Jetty offer more advanced capabilities, including six degrees of freedom (DOF) that enable greater flexibility and maneuverability. However, these solutions still rely on teleoperation, which can be challenging due to the complexity of controlling such a system manually.

[0005] US 2001 / 011841 A1 titled "Multiple jet hydrodemolition apparatus and method" discloses an apparatus for the hydrodemolition of concrete layer including a movable vehicle, a bed having a guideway extending transversely to a direction of movement of the vehicle, a nozzle assembly having a guide slidably engaging the guideway and a plurality of nozzles spaced apart in a direction transverse to the guideway, separate fluid flow controllers coupled between a pressurized source of fluid and respective nozzles, the nozzles being oriented to spray a fluid jet onto the concrete surface and means for moving the nozzle assembly along the guideway transversely across the direction of travel.

[0006] CA2879311A1 titled "SHIELDING SYSTEM FOR HYDRODEMOLITION APPARATUS" discloses a shield system for a hydrodemolition apparatus which has a loop whose ends are secured to opposite sides of a reciprocating nozzle carrier. The loop may be supported by a supporting structure surrounding the frame along which the nozzle carrier reciprocates. The loop overlays portions of one or more gaps through which one or more nozzles extend to work an underlying surface. The nozzles have access to the underlying surface through the gaps under the nozzle carrier, but the balance of the gaps are covered by the loop which is drawn or pushed by the nozzle carrier during reciprocation.

[0007] CN115726547A titled "Control method and device for automatically spraying wall body" discloses a control method and device for automatically spraying a wall. The method comprises obtaining a target wall image, recognizing a to-be-sprayed region in the target wall image, and decomposing the to-be-sprayed region into a plurality of sub-regions. A full-coverage spraying path in each sub-region is calculated, an adjacent graph is constructed according to the multiple sub-regions, and each sub-region is a node in the adjacent graph. The method further comprises performing path planning on each sub-region according to the adjacent graph to obtain an optimal path connecting an end point of each sub-region and a starting point of a next sub-region. The to-be-sprayed area is sprayed according to the optimal path and the full-coverage spraying path.

[0008] The existing state-of-the-art is characterized by several shortcomings. Firstly, current autonomous robots are unable to reliably perceive their environment or plan their own path, relying instead largely on manual intervention for navigation and control. This limits their ability to adapt to changing conditions or unexpected obstacles. Secondly, the machines' limited degrees of freedom make them difficult to use and maneuver in tight spaces or complex environments. This inflexibility can lead to reduced productivity and increased risk of damage or injury. Thirdly, many current solutions are not well-suited for processing hard-to-reach surfaces, requiring manual intervention or specialized equipment to access these areas. Fourthly, the machines' size and rigidity often make them difficult to fit into tight spaces, further limiting their effectiveness in certain applications. Lastly, most autonomous robots in this field require teleoperation by a worker, which can be time-consuming and labor-intensive. This reliance on human intervention limits the machines' autonomy and effectiveness.

[0009] In summary, while current solutions have shown promise in certain applications, they are often limited by their size, inflexibility, and lack of adaptability to complex surfaces or environments. The need for more advanced autonomous robots that can effectively process unstructured environments remains a pressing challenge in industries such as construction and concrete sanitation.

[0010] It is therefore a technical problem underlying the present invention to provide improved techniques for concrete deconstruction which are in particular more efficient.SUMMARY OF THE INVENTION

[0011] The problem is solved by the subject-matter defined in the independent claims. Advantageous modifications of embodiments of the invention are defined in the dependent claims as well as in the description and the figures.

[0012] According to a first aspect of the present invention, a method for concrete removal is provided. The method may be computer-implemented. The concrete removal may be performed using a robotic system, also referred to herein simply as robot or as mobile manipulator. The robotic system may be equipped with a high-pressure water nozzle.

[0013] The method may comprise a step a) of receiving, via a sensor system, first image data including an area of interest in the vicinity of the robotic system. The method may comprise a step b) of determining, based on the first image data, a concrete surface to be removed within the area of interest. The method may comprise a step c) of defining a coverage path for the high-pressure water nozzle to traverse the concrete surface to be removed. Said defining may include determining waypoints and / or corresponding nozzle orientations to achieve an optimized path that covers the concrete surface to be removed. The method may comprise a step d) of causing executing the concrete removal by the robotic system along the coverage path. Real-time distance sensing may be utilized to essentially maintain a predefined nozzle-to-surface distance, preferably in order to achieve a predetermined removal depth.

[0014] The term "robotic system" may be understood as an automated or semi-automated machine with one or more articulated arms, mobility platforms, or any combination thereof, specifically designed at least for concrete removal. The robotic system may include various sensors and actuator components configured to execute programmed tasks with precision.

[0015] The term "high-pressure water nozzle" may be understood as a nozzle capable of directing a jet of water at high pressures, which may be utilized as a means for concrete removal. The nozzle may be attached to the robotic system and may be controlled to move along a predefined path for effectively breaking up and removing concrete.

[0016] The term "sensor system" may be understood as at least one sensor, in particular a collection of sensors, such as cameras, LiDAR, ultrasonic sensors, and / or others, integrated in and / or being communicatively coupled to the robotic system. The sensor system may be responsible for capturing first image data and may also provide real-time distance measurements during the removal operation.

[0017] The term "image data" may be understood as a visual representation obtained from the sensor system that includes an area of interest in the vicinity of the robotic system. The data can be in the form of digital images and / or three-dimensional point clouds. It may be used to identify the concrete surface requiring removal.

[0018] The term "area of interest" may be understood as a specific zone within the vicinity of the robotic system where the robotic system is physically located. The area of interest includes a concrete surface that is targeted to be removed by executing the concrete removal process.

[0019] The term "concrete surface to be removed" may be understood as a particular section of concrete identified within the area of interest that has been determined to require removal.

[0020] The term "coverage path" may be understood as a strategically calculated route that the high-pressure water nozzle follows during the concrete removal process. The path ensures that the entire surface designated for removal is covered effectively. The term "waypoints" may include a set of spatial coordinates that define critical points along the coverage path which the high-pressure water nozzle follows.

[0021] The term "nozzle orientation" may be understood as an angular position of the high-pressure water nozzle relative to the concrete surface in three-dimensional space intended to maximize the efficiency of concrete removal. The nozzle orientation may be controlled by an arm of the robot system and / or by one or more hinge elements and / or the like.

[0022] The term "optimized path" may be understood as a particularly efficient route that has been calculated for the nozzle to traverse in order to remove essentially the entire predefined concrete surface. This may involve minimizing the time and / or water usage and / or energy required for the task, while achieving the desired level of concrete removal.

[0023] The term "executing concrete removal" indicates the operational phase where the robotic system carries out the actual physical process of removing the concrete using the high-pressure water nozzle, following the defined coverage path.

[0024] The term "real-time distance sensing" may be understood as an ongoing process of measuring the distance between the high-pressure water nozzle and the concrete surface during the removal operation, to ensure consistent removal depth and enhance overall precision.

[0025] The term "predefined nozzle-to-surface distance" may be understood as a distance that may be maintained between the nozzle and concrete surface to adequately achieve a predetermined level of removal depth without causing unnecessary damage to the underlying structure.

[0026] The term "predetermined removal depth" may be understood as a target depth to which the concrete needs to be removed, which can vary depending on the specific requirements of the application or construction project.

[0027] The combination of advanced imaging and sensor systems to receive first image data that includes the area of interest allows for precise identification and determination of the concrete surface needing removal. Consequently, the method averts the indiscriminate destruction of material, reducing waste and preserving structural integrity where necessary.

[0028] Further, the step of defining an optimized coverage path for the high-pressure water nozzle to traverse the concrete surface targeted for removal which includes the determination of waypoints and corresponding nozzle orientations, not only ensures thorough coverage of the concrete surface for removal but also significantly reduces the time and resources required to complete the task. As a result, the method enhances efficiency and can contribute to reduced operational costs.

[0029] Executing the concrete removal along the defined coverage path, the inventive process incorporates real-time distance sensing to maintain a predefined nozzle-to-surface distance consistently. This feature facilitates achieving a predetermined removal depth, which is essential for tasks requiring a high precision, such as in the preservation of underlying structural elements or when preparing surfaces for subsequent treatments or finishes. By maintaining consistent pressure and distance, the method minimizes variability in removal depth and avoids structural weaknesses that might arise from irregular surface treatment.

[0030] The use of real-time distance sensing also serves to accommodate for variations in the working environment, such as uneven surfaces or accessibility challenges. By adjusting the robotic system's operation dynamically, the method mitigates the risk of incomplete or excessive removal that could result in project delays or the need for costly rework.

[0031] Thus, an autonomous and efficient method for concrete removal using a robotic system is provided. This creates several further technical advantages: Better extraction selection, because the robot can either scan the environment or detect on its own what surface to process. The interface of the existing machines is usually difficult to operate, and teleoperating an end-effector does not work with good quality. The method can provide better extraction quality, due to the robot being able to hold the power of the water thrust better than a human. When looking at existing solutions, there are either workers doing the task manually, standard systems not being able to remove complex shapes due to their design, and systems which are teleoperated with a similar concept as ours. Humans cannot hold the water lance as accurate as our system due to the thrust of the water. Most known systems are not designed to be flexible but instead only for large, planar concrete removal. These systems therefore cannot extract concrete from round columns or corners for example. Compared to systems such as the Woma Jetty, embodiments of the invention can also be more accurate because the robot of the present invention is actuated with electricity and not with hydraulics, which are difficult to control. Current hydraulic arms are simply controlled by teleoperation, making control difficult and inefficient. Furthermore, with real-time distance measurement and control of the water lance to the surface, the robot can better adapt to the processing area than any current machines or humans. Iteratively improving results, because the robot can scan the wall after it processed the complete area, to see how deep it reached. If there is not enough concrete removed, the robot can again on its own plan a path to remove the rest of the concrete which was not done by the first sweep. Currently, this all must be done by manual labor. Bigger flexibility in the environment, due to the smaller robot with an arm with at least six DoF. The robot will be able to fit through doorways and can also work in environments where there is a lot of clutter and tight passages. With collision avoidance, we will be able to also let the robot work where current machines do not fit. Faster process, because the robot does not need to take a rest compared to humans.

[0032] Also, it can handle higher thrusts than humans, so that it can process surfaces faster. With autonomous movement, the robot furthermore will be able to place itself into the optimal positions for removing concrete, so the worker does not have to move the robot after every done extraction. Safer for humans, because manual labor can now be automated by a robotic solution. When working with current machines, not every surface can be processed, so manual labor sometimes still is necessary. Can be transported without a big truck, making the handling of the robot much easier. As the system is light enough to put in a van, no lorry license is required from a worker to bring the machine to the construction site. This makes the operation of the robot much easier to organize than with current systems. The robot can work with a small pump instead of container pump. This means that logistics are easier when not having to organize a space for the container to sit in, as well as being able to use normal water access instead of needing a water hydrant.

[0033] It may be provided that the method further comprises a step e) of, after executing the concrete removal, receiving, via the sensor system, second image data including the area of interest in the vicinity of the robotic system to detect any portions of the concrete surface to be removed that require additional concrete removal, in particular based on the predetermined removal depth, wherein optionally a before-after comparison map is generated based on the first and second image data to evaluate the effectiveness of concrete removal. This way, real-time feedback and evaluation of the concrete removal process can be provided.

[0034] It may be provided that the method further comprises a step f) of iteratively re-planning and executing supplementary coverage paths for portions identified as requiring additional concrete removal until a uniform removal depth is achieved throughout the concrete surface to be removed. This way, an iterative refinement process is provided that ensures a uniform removal depth across the entire concrete surface. By iteratively re-planning and executing supplementary coverage paths for portions identified as requiring additional concrete removal, the robot can achieve precise control, because the robot's advanced planning capabilities allow it to generate optimized supplementary coverage paths that precisely target areas requiring additional removal. Furthermore, rework may be minimized, because iterative refinement reduces the need for manual intervention and minimizes rework, resulting in a more efficient cleaning process with reduced labor costs. Lastly, uniformity may be ensured because by continuously monitoring and adjusting its performance, the robot ensures a consistent removal depth throughout the concrete surface, eliminating hotspots or areas that may require additional attention.

[0035] It may be provided that the sensor system includes a 3D LiDAR sensor and a wide field-of-view RGB camera, preferably being integrated into the robotic system. This way, accurate spatial mapping, enhanced object detection and improved scene understanding may be achieved.

[0036] It may be provided that step b) of determining the concrete surface to be removed within the area of interest includes: (i) detecting, based on the image data, painted markings within the area of interest, the painted markings at least partially indicating the concrete surface to be removed, wherein the painted markings may be provided by a user using a painting device configured to mark concrete surfaces in real world, optionally wherein an electronic user interface device visualizes the area of interest including the painted markings; By detecting painted markings that indicate the concrete surface to be removed, the robot can more accurately determine what areas need to be cleaned. The detection of painted markings provides real-time feedback on the environment, allowing the robot to adjust its behavior and make informed decisions about where to focus cleaning efforts. By following the painted markings, the robot can optimize its cleaning path and reduce rework or manual intervention.

[0037] In addition or alternatively, It may be provided that step b) of determining the concrete surface to be removed within the area of interest includes: (ii) receiving user input via the electronic user interface device, wherein determining the concrete surface to be removed is performed based on the user input, wherein preferably, the user is provided with a visualization of the area of interest by the electronic user interface device based on which the user input can be performed. Allowing users to provide input via an electronic interface device enables them to customize the cleaning process based on specific requirements or preferences. User-provided input can help ensure that the robot accurately identifies the concrete surface to be removed, reducing errors and improving overall performance. Providing a visualization of the area of interest via an electronic interface device allows users to better understand the cleaning process and make informed decisions about where to focus their efforts. Therefore, in other words, areas of interest can be marked by a digital pen by tracing the circumference of the current area to process, or by marking surfaces on the electronic interface device which may be a control tablet.

[0038] It may be provided that the method further comprises visualizing, on the electronic user interface device, the defined coverage path for approval and / or refinement by the user. By providing a clear visualization of the defined coverage path, users can better understand how their input will be translated into action. Allowing users to review and refine the defined coverage path enables them to tailor the cleaning process to specific requirements or preferences. Visualizing the defined coverage path helps prevent mistakes by allowing users to identify potential issues before the robot begins cleaning. By refining the defined coverage path, users can optimize the cleaning process and reduce rework or manual intervention.

[0039] It may be provided that the method further comprises transitioning the robotic system to a safe state under certain safety conditions, received from a plurality of safety sensors, and as decided by a programmed safety system within the robotic system, optionally wherein the safe state includes stopping movement and / or stopping processing of the robotic system. The ability for the robot to automatically transition into a safe state in response to changing environmental or sensor data ensures that potential hazards are mitigated, reducing the risk of accidents or damage. By integrating multiple safety sensors and a programmed safety system, the robotic system can quickly respond to unexpected situations, such as obstacles or changes in its environment. The ability for the robot to transition into different safe states (e.g., stopping movement and / or processing) allows it to adapt to various scenarios, ensuring continued operation within defined safety parameters. By automating the process of transitioning to a safe state, the robotic system reduces the need for human intervention or manual override, simplifying overall control and reducing potential errors It may be provided that real-time distance sensing is performed using ultrasonic sensors. Ultrasonic sensors provide accurate measurements of distances, even in environments with varying lighting conditions or reflective surfaces. Real-time distance sensing enables the robotic system to receive continuous updates on its surroundings, allowing for more precise navigation and obstacle avoidance. Ultrasonic sensors can operate effectively over a wide range of distances (typically up to 10-15 meters), making them suitable for various applications. Ultrasonic sensors are less susceptible to electromagnetic interference or radio frequency interference, ensuring reliable distance sensing in environments with high levels of electrical activity.

[0040] It may be provided that method further comprises adjusting water pressure and / or flow rate based on the real-time distance sensing and / or the type of concrete surface encountered during the concrete removal. By adjusting water pressure and / or flow rate in response to changing distances or surface types, the robotic system can optimize its cleaning performance, reducing rework or manual intervention. Real-time adjustments enable the robot to tailor its cleaning approach to specific concrete surfaces, ensuring effective removal of dirt, grime, and other contaminants without damaging the surface. By optimizing water pressure and flow rate based on real-time sensing data, the robotic system can minimize water usage while still achieving effective cleaning results. The ability to adjust water pressure and / or flow rate in response to changing conditions enables the robot to adapt to various concrete surfaces, including those with different textures or porosities.

[0041] It may be provided that the execution according to step d) is performed autonomously by the robot system, preferably offline and / or without additional inputs from the user until completion of the coverage path or upon encountering a safety condition that triggers safe state transition. By automating this step, the robotic system can complete its cleaning task independently, reducing the need for human intervention and minimizing downtime. Autonomous operation reduces the risk of accidents or injuries caused by human error or distraction during the cleaning process. Offline autonomous operation enables the robot to adapt to changing environmental conditions or unexpected obstacles without requiring user input or manual override. By automating the step, the robotic system can reduce labor costs associated with manual intervention and minimize the need for human supervision.

[0042] It may be provided that the robotic system includes a robotic arm where the high-pressure water nozzle is located, the robotic arm having at least six degrees of freedom to manipulate the orientation of the high-pressure water nozzle during traversal of the coverage path. The ability to manipulate the orientation of the high-pressure water nozzle through multiple DOFs allows for more precise control and adaptability during traversal of the coverage path, enabling the robot to reach complex or hard-to-access areas. With a robotic arm capable of manipulating its orientation in six dimensions (e.g., x, y, z, roll, pitch, yaw), the system can achieve better coverage of the concrete surface by adjusting the nozzle's angle and position to ensure thorough cleaning. The increased flexibility provided by the robotic arm enables it to reach areas that would be difficult or impossible for a fixed-nozzle robot to access, such as corners, edges, or recessed surfaces. By precisely controlling the orientation of the high-pressure water nozzle, the system can minimize scratches and damage on the concrete surface, ensuring a smoother finish and reducing the risk of defects. The ability to manipulate the nozzle's orientation allows for more precise control over the cleaning process, enabling the robot to avoid sensitive areas or obstacles while maintaining a safe distance from them.

[0043] Based on the received first image data, a first point cloud of the area of interest, preferably a first colored point cloud of the area of interest, may be generated. The same may be performed for the second image data. For example, steps c), d), e), f), (i) and / or (ii) may be performed on the basis of such point cloud.

[0044] According to another aspect of the invention, a robot system for concrete removal is provided. The robot system is configured to perform any of the methods described herein. It may be provided that the robot system comprises a mobile base equipped with a vehicle platform for navigating various terrain types, the vehicle platform preferably being a tracked vehicle platform. It may be provided that the robotic system comprises a robotic arm coupled to the mobile base. The robotic arm may provide at least six degrees of freedom for maneuvering an attached high-pressure water nozzle designed to perform removal of concrete surfaces, wherein preferably the high-pressure water nozzle is interchangeably attached to the robotic arm, allowing for the substitution of the nozzle with other end-effectors for performing additional functions beyond concrete removal. It may be provided that the robotic system comprises a sensor system configured to scan an area of interest in the vicinity of the robot system. It may be provided that the robotic system comprises a processing unit communicatively coupled with the robotic arm and the sensor system. It may be provided that the robotic system comprises an electronic user interface device or a communication interface configured to communicate with an electronic user interface device. Optionally, it may be provided that the robotic system comprises a safety system equipped with safety sensors including radars for detecting humans and / or sensors for staircase fall protection.

[0045] According to another aspect of the invention, a data processing apparatus is provided. The data processing apparatus is configured to perform any of the methods described herein.

[0046] According to another aspect of the invention, a computer program or a computer-readable medium having stored thereon a computer program is provided. The computer program comprises instructions which, when the program is executed by a computer, preferably by a data processing apparatus as described above, more preferably by a robotic system as described above, cause the robotic system to carry out any of the methods described herein

[0047] All features, technical implementation details and advantages described with respect to any one of the aspects of the present invention described herein are self-evidently mutatis mutandis applicable for any one of the other aspects of the present invention and vice versa.BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The disclosure may be better understood by reference to the following drawings: Fig. 1:A schematic block diagram of a robotic system according to embodiments of the present invention. Fig. 2:A flow chart of a process for operating the robotic system according to embodiments of the present invention. Fig. 3:A photograph of marked and deteriorated concrete which needs to be removed according to embodiments of the present invention. Fig. 4:Images illustrating a process of finding a marked area and determining a trajectory for the autonomous robot to follow according to embodiments of the present invention. Fig. 5:A pseudocode listing of an algorithm for detecting 3D waypoints based on an image according to embodiments of the present invention. Fig. 6:A visualization of how to find waypoints for coverage traversal of a shape according to embodiments of the present invention. Fig. 7:A pseudocode listing of an algorithm for finding the horizontal pixels needed for shifting a vertical trajectory by a given step size in 3D according to embodiments of the present invention. DETAILED DESCRIPTION

[0049] Fig. 1 shows a schematic block diagram of a robotic system 102, also referred to as mobile manipulator or simply robot herein, according to an exemplary embodiment of the invention. The mobile manipulator 102 comprises multiple building blocks. The mobile basis 114 enables the robot 102 to move in its environment. The mobile basis 114 is preferably a tracked vehicle enabling the traversal of various kinds of ground structure. On the mobile basis 114, the rest of the robot 102 can be mounted in a modular fashion, to enable the usage of different robotic arms, end-effectors, as well as utility trunks for the robot 102.

[0050] The safety system 118 can be provided on a programmable logic controller (PLC) configured to determine if the robot 102 needs to go to a safe state to not endanger its environment. The safe state may comprise simply stopping the robot 102 so that it does not move. The safety system 118 is fed by a sensor system 120 including one or more sensors configured to determine the current safety state.

[0051] The sensors 120 of the robot 102 may comprise one or more depth cameras 122, one or more LiDARs 122, one or more RGB cameras 124 or any combination thereof. In a preferred embodiment, the sensors 120 the robot 102 comprises a 3D LiDAR 122 combined with wide-FoV RGB cameras 124 to collect a colored point cloud. The point cloud can be used for determining what area the robot 102 should process. The sensory 120 may also comprise safety sensors such as radars capable of scanning for humans in the environment. These sensors can be used for determining if the robot is in a safe state or not. Furthermore, there may be staircase fall protection sensors, distance sensors for the end-effector, ultrasonic sensors, or any other combination of these sensors.

[0052] The 3D LiDAR as well as the end-effector may be mounted on the robot arm 112. This way, the robot 112 can act as a manipulator in the environment, where it is able to autonomously scan the surroundings and plan paths on its own to process surfaces. The manipulator will be exchangeable to enable the usage of different arms for different use cases.

[0053] The computer 116, also referred to as processing unit herein, of the robot 102 is configured to gather sensor inputs and to calculate according to the current state the path the robot 102 should take. This may involve especially how coverage path planning is done for surface traversal.

[0054] The high-pressure water nozzle 110 is the end-effector in the illustrated embodiment, capable of removing concrete by ejecting water onto the surface in front. This end-effector can also be exchanged, so that the robot 102 can take over other tasks such as painting or gripping things.

[0055] Outside of the mobile manipulator system 102 is a water pump 104 providing pressurized water to the robot 102. The water may be fed through a high-pressure capable water hose to the robot 102.

[0056] A worker 106 may control the robotic system 102, e.g., with a control tablet 108 or another electronic user interface device.

[0057] Fig. 2 shows a flow chart illustrating a process of operating the robotic system 102. In certain embodiments, there is a worker 106 controlling the system to have a human-in-the-loop. In step 202, the worker 106 first steers the robot 102 to the wall where there is concrete to be removed. This may include unloading the robot 102 from a truck it is transported in, connecting it to the water hose 104 and electricity if needed, and steering it into the garage, for example.

[0058] In step 204, the robot 102 scans the current environment and displays it on the tablet 108 the worker 106 holds. The environmental scanning process must be done carefully in order to not crash into the surroundings, as these are unknown beforehand.

[0059] In step 206, the worker 106 can then digitally define an area on the control tablet 108 to indicate the area on which the robot 102 should work on. This can be done by either pressing on a wall and fully selecting it, or by marking a specific area and only doing path planning for this. Also, automatic detection of construction spray can be done so that only the detected area needs to be selected by one touch on the display.

[0060] In step 208, the robot 102 plans and visualizes the path it calculated. This is important to show the supervising worker 106 what the robot 102 is going to do. The worker 106 is given the chance to check the path and start the execution in step 210. If the planned path is not correct, the selection procedure of the wall can be repeated, and another path can be planned in step 208. If the surface is traversed correctly, the worker 106 can approve the procedure and it gets sent to the robot 102 for execution in step 214.

[0061] The high-pressure water concrete removal is then done offline by the robot 102. This means that during the procedure, the path is not recalculated, only distance sensors may be used to determine if the end-effector is too close or too far away from the wall. This is important because the nozzle should not crash into the wall to prevent damage to the hardware, but being too far away deteriorates the extraction quality.

[0062] In the following a variation of a concrete removal process according to an exemplary embodiment of the invention is explained from the point of view of the worker 106. 1. Construction worker 106 takes tablet control 108 and drives robot 102 out of van to place it for preparation. 2. Worker 106 takes water hose and electricity cable and attaches it to robot 102 to make it ready for concrete removal. 3. Worker 106 takes tablet 108 and drives robot 102 into construction site to place it in front of a surface which needs concrete removal. 4. Worker 106 presses "Scan" on tablet 108, so that the robot 102 scans its environment for creating a map. 5. Worker 106 a) Selects an area on the tablet 108 which was detected by the robot 102. b) Takes a digital marking device and circles the area to process. c) Circles the area to process on the tablet 108. to indicate the robot 102 what surface it should plan a trajectory on. 6. The robot 102 takes the marked surface in the form of a point cloud and plans a trajectory to traverse the complete area. 7. The robot 102 displays the trajectory it planned on the tablet 108 so that the worker 106 can check it. 8. The worker 106 checks the trajectory for any errors and either accepts it or tells the robot 102 to replan. For replanning, go back to step 5. 9. If the trajectory is accepted, the robot 102 receives the start command and begins to increase water pressure and flow, to start concrete removal. The worker 106 is at a safe distance and monitors the process. 10. After the robot 102 finishes the traversal, it moves back to the home position and waits for the moisture in the air and the dust to settle. 11. As soon as the vision is clear again, the robot 102 rescans the processed area to produce a before-after comparison. 12. If there are parts of the area which have not been extracted deep enough, the robot 102 plans another trajectory over only these parts to reprocess the surface to ensure equal extraction depth. Go back to step 7. 13. If the concrete extraction has been successful in every part of the area, the worker 106 takes the robot control tablet 108 and moves the robot 102 to the next position for further concrete removal tasks.

[0063] In the following, an exemplary implementation of the above concepts in accordance with an exemplary embodiment of the invention is described. The autonomous concrete removal robot 102 of this embodiment is configured to receive a point cloud and RGB camera information as inputs. The robot 102 is configured to recognize a marked shape outlined by an expert 106 with spray paint. An example is shown in Fig. 3. The robot 102 is configured to automatically assign waypoints onto the boundaries of the area.

[0064] The used algorithm enables the robot 102 to first detect the marked figure on a wall in front of the camera. After finding the shape, waypoints are determined and the corresponding normal orientations are calculated. Between each vertical stride, the horizontal step size is determined according to a specified distance. After finding all waypoints, the robot 102 is supposed to traverse them. During the entire operation, the robot 102 must avoid collisions with the wall and ground.

[0065] To do this, the sensor data is processed. The image becomes an OpenCV image, and the point cloud gets converted to an array of three dimensions, where the coordinates in space get saved at the corresponding pixel values in the array. The remaining process works with the image data gained. In the illustrated embodiment, the marked area is indicated by red spray paint. First, a Boolean mask is created, indicating the red pixels with true and the rest with false. Two arrays are specified that contain the lower and upper bounds of the HSV values we determine as marked. As the red color can be found on both edges of the hue spectrum, this is done twice, once for the lower edge and once for the higher edge of the range. A mask is then created by first finding the lower and higher masks and adding them. The process is illustrated in Fig. 4.

[0066] To set waypoints on the boundary of the marked area, the contour of the image is calculated, as we need it to fill the mask, by using the OpenCV function findContours(), where we need to give the mask as an argument. This contour is then filled with the function fillPoly(). Now that the filled area is determined from the image taken with the camera, the algorithm can start trying to set the waypoints to traverse the figure. One method for setting waypoints is described in the algorithm shown in Fig. 5.

[0067] In certain embodiment, finding the waypoints is performed as follows: First, find the outer bounding box of the mask. Then start inside this area in the upper left corner and loop vertically through the pixels. When a pixel is inside the shape, but its neighbor pixel above or below is outside, it is on the boundary of the marked area. We append the boundary pixel to a list and continue our loop. This search continues until we arrive at the y-position of the lowest extremum of the shape. Here we take a horizontal step to the right, specified by a step size with another method we will explain next. We then set our y-position to the highest point of the bounding box and repeat the described process until we take enough horizontal steps to arrive at the right extremum. The process is visualized in Fig. 6.

[0068] It is essential to have the correct step size, as we want even coverage of the water jet on the wall. To take a fitting step size, we implemented the method horizontal_step() depicted in Fig. 7. It receives the step size and current pixel coordinates on the image from which we want to take the step. The goal is to take a pixel step to the right until we reach the given spatial step size. The loop behaves as follows: If the covered distance is still less than the given step size, we move one pixel to the right, increasing the pixel's horizontal position. At this point, we calculate the spatial distance between the current pixel position and the position to the left. This distance gets added to the covered distance every time we pass the loop. Important to note is that we should not calculate the Euclidean distance in 3D space as pixels on the same pixel row might have different heights in space. We are only interested in the horizontal distance traveled. We should ignore the difference in the height of the coordinates and only calculate the Euclidean distance in the x - y plane.

[0069] As soon as the waypoints are determined, a trajectory can be calculated. This can be done by standard measures such as Rapidly Exploring Random Trees (RRT). While calculating the trajectory every robot pose can also be assessed for collisions with itself and the environment. The result is a robot trajectory which follows the given waypoints in order, without colliding with the environment.

[0070] Now that the robot was able to autonomously find a trajectory for concrete removal by scanning the environment and filtering for the market shape, the execution can be started. This may simply be done by giving the robot the respective trajectory and instructing it to follow the path.

[0071] Although some aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the corresponding method, where a block or device corresponds to a method step or a feature of a method step. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus.

[0072] Embodiments of the invention may be implemented on a computer system. The computer system may be a local computer device (e.g. personal computer, laptop, tablet computer or mobile phone) with one or more processors and one or more storage devices or may be a distributed computer system (e.g. a cloud computing system with one or more processors and one or more storage devices distributed at various locations, for example, at a local client and / or one or more remote server farms and / or data centers). The computer system may comprise any circuit or combination of circuits. In one embodiment, the computer system may include one or more processors which can be of any type. As used herein, processor may mean any type of computational circuit, such as but not limited to a microprocessor, a microcontroller, a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a graphics processor, a digital signal processor (DSP), multiple core processor, a field programmable gate array (FPGA), or any other type of processor or processing circuit. Other types of circuits that may be included in the computer system may be a custom circuit, an application-specific integrated circuit (ASIC), or the like, such as, for example, one or more circuits (such as a communication circuit) for use in wireless devices like mobile telephones, tablet computers, laptop computers, two-way radios, and similar electronic systems. The computer system may include one or more storage devices, which may include one or more memory elements suitable to the particular application, such as a main memory in the form of random access memory (RAM), one or more hard drives, and / or one or more drives that handle removable media such as compact disks (CD), flash memory cards, digital video disk (DVD), and the like. The computer system may also include a display device, one or more speakers, and a keyboard and / or controller, which can include a mouse, trackball, touch screen, voice-recognition device, or any other device that permits a system user to input information into and receive information from the computer system.

[0073] Some or all of the method steps may be executed by (or using) a hardware apparatus, like for example, a processor, a microprocessor, a programmable computer or an electronic circuit. In some embodiments, some one or more of the most important method steps may be executed by such an apparatus.

[0074] Depending on certain implementation requirements, embodiments of the invention can be implemented in hardware or in software. The implementation can be performed using a non-transitory storage medium such as a digital storage medium, for example a floppy disc, a DVD, a Blu-Ray, a CD, a ROM, a PROM, and EPROM, an EEPROM or a FLASH memory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable.

[0075] Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.

[0076] Generally, embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer. The program code may, for example, be stored on a machine readable carrier.

[0077] Other embodiments comprise the computer program for performing one of the methods described herein, stored on a machine readable carrier.

[0078] In other words, an embodiment of the present invention is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

[0079] A further embodiment of the present invention is, therefore, a storage medium (or a data carrier, or a computer-readable medium) comprising, stored thereon, the computer program for performing one of the methods described herein when it is performed by a processor. The data carrier, the digital storage medium or the recorded medium are typically tangible and / or non-transitory. A further embodiment of the present invention is an apparatus as described herein comprising a processor and the storage medium.

[0080] A further embodiment of the invention is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein. The data stream or the sequence of signals may, for example, be configured to be transferred via a data communication connection, for example, via the internet.

[0081] A further embodiment comprises a processing means, for example, a computer or a programmable logic device, configured to, or adapted to, perform one of the methods described herein.

[0082] A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0083] A further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of the methods described herein to a receiver. The receiver may, for example, be a computer, a mobile device, a memory device or the like. The apparatus or system may, for example, comprise a file server for transferring the computer program to the receiver.

[0084] In some embodiments, a programmable logic device (for example, a field programmable gate array) may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

Examples

Embodiment Construction

[0049]Fig. 1 shows a schematic block diagram of a robotic system 102, also referred to as mobile manipulator or simply robot herein, according to an exemplary embodiment of the invention. The mobile manipulator 102 comprises multiple building blocks. The mobile basis 114 enables the robot 102 to move in its environment. The mobile basis 114 is preferably a tracked vehicle enabling the traversal of various kinds of ground structure. On the mobile basis 114, the rest of the robot 102 can be mounted in a modular fashion, to enable the usage of different robotic arms, end-effectors, as well as utility trunks for the robot 102.

[0050]The safety system 118 can be provided on a programmable logic controller (PLC) configured to determine if the robot 102 needs to go to a safe state to not endanger its environment. The safe state may comprise simply stopping the robot 102 so that it does not move. The safety system 118 is fed by a sensor system 120 including one or more sensors configured to ...

Claims

1. A computer-implemented method for concrete removal using a robotic system (102) equipped with a high-pressure water nozzle (110), the method comprising the following steps: a) receiving, via a sensor system (120), first image data including an area of interest in the vicinity of the robotic system (102); b) determining, based on the first image data, a concrete surface to be removed within the area of interest; c) defining (208) a coverage path for the high-pressure water nozzle (110) to traverse the concrete surface to be removed, wherein said defining includes determining waypoints and / or corresponding nozzle orientations to achieve an optimized path that covers the concrete surface to be removed; and d) causing executing (214) the concrete removal by the robotic system (102) along the coverage path, wherein real-time distance sensing is utilized to essentially maintain a predefined nozzle-to-surface distance, preferably in order to achieve a predetermined removal depth.

2. The computer-implemented method of claim 1, further comprising: e) after executing (214) the concrete removal, receiving, via the sensor system (120), second image data including the area of interest in the vicinity of the robotic system (102) to detect any portions of the concrete surface to be removed that require additional concrete removal, in particular based on the predetermined removal depth, wherein optionally a before-after comparison map is generated based on the first and second image data to evaluate the effectiveness of concrete removal.

3. The computer-implemented method of claim 2, further comprising: f) iteratively re-planning and executing supplementary coverage paths for portions identified as requiring additional concrete removal until a uniform removal depth is achieved throughout the concrete surface to be removed.

4. The computer-implemented method of any one of the preceding claims, wherein the at sensor system (120) includes a 3D LiDAR sensor (122) and a wide field-of-view RGB camera (124), preferably being integrated into the robotic system (102).

5. The computer-implemented method of any one of the preceding claims, wherein step b) of determining the concrete surface to be removed within the area of interest includes: (i) detecting, based on the image data, painted markings within the area of interest, the painted markings at least partially indicating the concrete surface to be removed, wherein the painted markings may be provided by a user (106) using a painting device configured to mark concrete surfaces in real world, optionally wherein an electronic user interface device (108) visualizes the area of interest including the painted markings; and / or (ii) receiving user input via the electronic user interface device (108), wherein determining the concrete surface to be removed is performed based on the user input, wherein preferably, the user (106) is provided with a visualization of the area of interest by the electronic user interface device (108) based on which the user input can be performed.

6. The computer-implemented method of any one of the preceding claims, comprising visualizing, on the electronic user interface device (108), the defined coverage path for approval and / or refinement by the user (106).

7. The computer-implemented method of any one of the preceding claims, comprising transitioning the robotic system (102) to a safe state under certain safety conditions, received from a plurality of safety sensors, and as decided by a programmed safety system (118) within the robotic system (102), optionally wherein the safe state includes stopping movement and / or stopping processing of the robotic system (102).

8. The computer-implemented method of any one of the preceding claims, wherein real-time distance sensing is performed using ultrasonic sensors.

9. The computer-implemented method of claim 8, further comprising adjusting water pressure and / or flow rate based on the real-time distance sensing and / or the type of concrete surface encountered during the concrete removal.

10. The computer-implemented method of any one of the preceding claims, wherein the execution according to step d) is performed autonomously by the robotic system (102), preferably offline and / or without additional inputs from the user (106) until completion of the coverage path or upon encountering a safety condition that triggers safe state transition.

11. The computer-implemented method of any one of the preceding claims, wherein the robotic system (102) includes a robotic arm (112) where the high-pressure water nozzle (110) is located, the robotic arm (112) having at least six degrees of freedom to manipulate the orientation of the high-pressure water nozzle (110) during traversal of the coverage path.

12. A robotic system (102) for concrete removal configured to perform a method according to any one of claims 1-11.

13. The robotic system (102) according to claim 12, comprising: a) a mobile base (114) equipped with a vehicle platform for navigating various terrain types, the vehicle platform preferably being a tracked vehicle platform; b) a robotic arm (112) coupled to the mobile base (114), providing at least six degrees of freedom for maneuvering an attached high-pressure water nozzle (110) designed to perform removal of concrete surfaces, wherein preferably the high-pressure water nozzle (110) is interchangeably attached to the robotic arm (112), allowing for the substitution of the nozzle with other end-effectors for performing additional functions beyond concrete removal; c) a sensor system (120) configured to scan an area of interest in the vicinity of the robotic system (102); d) a processing unit (116) communicatively coupled with the robotic arm (112) and the sensor system (120); e) an electronic user interface device (108) or a communication interface configured to communicate with an electronic user interface device (108); and f) optionally, a safety system (118) equipped with safety sensors including radars for detecting humans and / or sensors for staircase fall protection.

14. A data processing apparatus being configured to perform a method according to any one of claims 1-11.

15. A computer program or a computer-readable medium having stored thereon a computer program, the computer program comprising instructions which, when the program is executed by a computer, preferably by a data processing apparatus according to claim 14, more preferably by a robotic system (102) according to claim 12 or 13, cause the robotic system (102) to carry out a method of any one of claims 1 to 11.

Images