A boom anti-collision detection method, electronic equipment and multi-boom device

Abstract

The application discloses an arm support anti-collision detection method, comprising the following steps: acquiring first point cloud data of the surrounding environment of a multi-arm support device; detecting obstacles in the surrounding environment according to the first point cloud data and establishing a first circumscribed bounding box of the obstacles; performing first collision detection according to the first circumscribed bounding box and a second circumscribed bounding box of an arm support in the multi-arm support device; in the case that the result of the first collision detection indicates that there is a first obstacle that will collide with the first arm support, performing second collision detection in the coordinate system corresponding to the first arm support based on the point cloud data corresponding to the first obstacle in the first point cloud data; and in the case that the result of the second collision detection indicates that the first obstacle will collide with the first arm support, generating anti-collision alarm information. The application can automatically perform collision detection, and after one collision detection based on the point cloud data of the surrounding environment, secondary collision detection is performed according to the point cloud data of the obstacle that will collide, so that the detection result is more accurate.

Description

Technical Field

[0001] This invention relates to the field of automatic control, and in particular to a method for detecting collisions with a boom, an electronic device, and a multi-boom device. Background Technology

[0002] Currently, the operation of multi-joint boom engineering equipment such as pump trucks and truck cranes mainly relies on the coordinated work of multiple personnel to adjust the boom. These machines have long booms with multiple positions, and often operate at heights. When relying on the naked eye to observe surrounding obstacles, the limited field of vision makes it impossible to fully observe the entire area around the boom. This increases the risk of collisions between the boom and obstacles during boom adjustments, leading to safety accidents. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a boom collision detection method, electronic device and multi-boom device, which can automatically perform collision detection, and after performing a collision detection based on the point cloud data of the surrounding environment, a second collision detection is performed based on the point cloud data of the obstacle that is in collision, so that the detection result is more accurate.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0005] A boom collision avoidance detection method, applied to multi-boom equipment, the method comprising:

[0006] S1 acquires the first point cloud data of the environment surrounding the multi-arm device;

[0007] S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes a first bounding box for the obstacles;

[0008] S3 performs a first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom device;

[0009] S4 If the result of the first collision detection indicates that there is a first obstacle and the first boom colliding, S4 performs a second collision detection in the coordinate system corresponding to the first boom based on the second point cloud data, wherein the obstacle in the surrounding environment includes the first obstacle, the multi-boom device includes the first boom, and the second point cloud data is the point cloud data in the first point cloud data corresponding to the first obstacle.

[0010] S5 generates a collision avoidance alarm if the result of the second collision detection indicates that the first obstacle has collided with the first boom.

[0011] Optionally, in step S4, when performing a second collision detection based on the second point cloud data in the coordinate system corresponding to the first boom, the method includes:

[0012] Voxel subdivision is performed on the 3D point cloud corresponding to the second point cloud data;

[0013] Based on the results of the voxel subdivision, multiple third bounding boxes corresponding to the first obstacle are established;

[0014] The second collision detection is performed based on the coordinates of the third external bounding box in the coordinate system corresponding to the first boom.

[0015] Optionally, in S3, during the first collision detection based on the first bounding box and the second bounding box, the method includes:

[0016] The first external bounding box and the second external bounding box perform the first collision detection in the coordinate system of the multi-arm equipment.

[0017] Optionally, after S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes a first bounding box for the obstacles, the method includes:

[0018] Obstacle avoidance path planning is performed based on the first external bounding box.

[0019] Optionally, the second external bounding box includes a first sub-bounding box of the coordinate system corresponding to the multi-arm device and a second sub-bounding box of the coordinate system corresponding to the first arm;

[0020] In S3, during the first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom device, the method includes:

[0021] Determine whether the first outer bounding box intersects with the first sub-bounding box;

[0022] If the first outer bounding box and the first sub-bounding box do not intersect, the result of the first collision detection indicates that there is no first obstacle and the first boom that collided;

[0023] If the first outer bounding box intersects with the first sub-bounding box, the distance between the first outer bounding box and the second sub-bounding box is used to determine whether the conditions for a collision are met.

[0024] If the conditions for a collision are met, the first obstacle and the first boom that collided are identified, and the result of the first collision detection indicates that there is a first obstacle and the first boom that collided.

[0025] If the conditions for a collision are not met, the result of the first collision detection indicates that there is no first obstacle and first boom that will collide.

[0026] Optionally, in acquiring the first point cloud data of the environment surrounding the multi-arm device in S1, the method includes:

[0027] When the multi-arm device rotates to the left, the point cloud data corresponding to the left side of the arm obtained from the lidar data is used as the first point cloud data;

[0028] When the multi-arm device rotates to the right, the point cloud data corresponding to the right side of the arm obtained from the lidar data is used as the first point cloud data;

[0029] When the multi-arm device undergoes a luffing action, the point cloud data corresponding to a preset range around the central axis of the arm obtained from the lidar data is used as the first point cloud data.

[0030] Optionally, in step S2, the method includes detecting obstacles in the surrounding environment based on the first point cloud data and establishing a first bounding box for the obstacles, the method further includes:

[0031] The three-dimensional point cloud corresponding to the first point cloud data is divided into voxel subdivisions to establish multiple first grid spaces;

[0032] Cluster analysis is performed on the point cloud within the first grid space to detect obstacles in the surrounding environment and determine the first bounding box of the obstacles.

[0033] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of any of the methods described above.

[0034] This application also provides a multi-arm device, including a lidar and a controller, wherein the lidar is used to collect data to obtain the first point cloud data, and the controller executes a computer program to implement the steps of any of the methods described above.

[0035] This application also provides a computer-readable storage medium storing a computer program that performs the steps of any of the methods described above.

[0036] As described above, this invention discloses a boom collision avoidance detection method, comprising: acquiring first point cloud data of the environment surrounding a multi-boom device; detecting obstacles in the surrounding environment based on the first point cloud data and establishing a first bounding box for the obstacles; performing a first collision detection based on the first bounding box and a second bounding box for the boom in the multi-boom device; if the result of the first collision detection indicates the existence of a first obstacle and a first boom colliding, performing a second collision detection based on the point cloud data corresponding to the first obstacle in the first point cloud data, in the coordinate system corresponding to the first boom; and generating a collision avoidance alarm information if the result of the second collision detection indicates a collision between the first obstacle and the first boom. This invention can automatically perform collision detection, and by performing a first collision detection based on the point cloud data of the surrounding environment, and then performing a second collision detection based on the point cloud data of the obstacle colliding, the detection results are more accurate. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating the boom collision avoidance detection method provided by the present invention.

[0039] Figure 2 This is another flowchart illustrating the boom collision avoidance detection method provided by the present invention.

[0040] Figure 3 A schematic diagram of the structure of the multi-arm device provided by the present invention.

[0041] Figure 4 This is a schematic diagram of the hardware layout on the multi-arm device provided by the present invention. Detailed Implementation

[0042] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. Based on the description of the present invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present invention.

[0043] In the description of this invention, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0044] The terms “first,” “second,” “third,” etc., are used merely to distinguish numerical values ​​or elements with similar properties, rather than to indicate or imply relative importance or a specific order.

[0045] The terms “include,” “comprising,” or any other variation thereof are intended to cover non-exclusive inclusion, which includes not only the elements listed but also other elements not expressly listed.

[0046] Figure 1 This is a flowchart illustrating the boom collision avoidance detection method provided by the present invention. Figure 1 As shown, a boom collision avoidance detection method, applied to multi-boom equipment, includes:

[0047] S1 acquires the first point cloud data of the environment surrounding the multi-arm equipment;

[0048] S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes the first bounding box of the obstacles;

[0049] S3 performs a first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom equipment;

[0050] S4 If the result of the first collision detection indicates that there is a first obstacle and the first boom that collided, S4 performs a second collision detection in the coordinate system corresponding to the first boom based on the second point cloud data. The obstacles in the surrounding environment include the first obstacle, the multi-boom equipment includes the first boom, and the second point cloud data is the point cloud data corresponding to the first obstacle in the first point cloud data.

[0051] S5 generates a collision avoidance alarm if the result of the second collision detection indicates that the first obstacle has collided with the first boom.

[0052] Through the above method, this application realizes automatic collision detection. Furthermore, after performing a collision detection based on the point cloud data of the surrounding environment, a second collision detection is performed based on the point cloud data of the obstacles that are colliding. There are two levels of detection process, coarse and fine, which results in higher detection accuracy and more accurate results.

[0053] The following combination Figure 1 and Figure 2The boom collision avoidance detection method of this embodiment will be introduced.

[0054] In step S1, a lidar is used to collect data to obtain the first point cloud data. Please refer to [link / reference needed]. Figure 3 LiDAR 702 can be installed at multiple key locations on the multi-section boom 707 of the pump truck. For example, when the boom has six sections, LiDAR can be installed at sections 1, 3, and 5. This ensures full coverage of the boom's sensing area under typical working conditions with minimal LiDAR installations, effectively reducing costs. It should be understood that in other embodiments, LiDAR can also be installed at sections 2, 4, and 6.

[0055] Optionally, in acquiring the first point cloud data of the environment surrounding the multi-arm equipment in S1, the method includes:

[0056] When the multi-arm equipment rotates to the left, the point cloud data corresponding to the left side of the arm obtained from the lidar data is used as the first point cloud data;

[0057] When the multi-arm equipment rotates to the right, the point cloud data corresponding to the right side of the arm obtained from the lidar data is used as the first point cloud data;

[0058] When the multi-arm equipment undergoes a luffing action, the point cloud data corresponding to the preset range around the central axis of the arm obtained from the lidar data is used as the first point cloud data.

[0059] Therefore, based on the boom's operating conditions, an effective data acquisition strategy for the LiDAR is formulated to obtain an effective region of interest for obstacle detection, thereby improving computational efficiency. The preset range around the boom's central axis can be taken as the range [-n, n] around the boom's central axis, where n is 0.1-5 meters, preferably 1 meter. For example, if n is 1 meter, then the current region of interest is a cylinder with a radius of 1 meter centered on the boom's central axis.

[0060] Optionally, in S2, obstacles in the surrounding environment are detected based on the first point cloud data, and a first bounding box for the obstacles is established. The method includes:

[0061] The three-dimensional point cloud corresponding to the first point cloud data is divided into voxel subdivisions to establish multiple first grid spaces;

[0062] Cluster analysis is performed on the point cloud within the first grid space to detect obstacles in the surrounding environment and determine the first bounding box of the obstacles.

[0063] In this embodiment, the 3D point cloud corresponding to the first point cloud data is divided into voxel subdivisions along either the x or y axis, with a subdivision interval of a first preset distance. In this embodiment, the first preset distance is 1 meter, thereby forming W first grid spaces (W = X / s or Y / s, where X and Y represent the side lengths of the 3D point cloud corresponding to the first point cloud data along the x and y axes, respectively). In other embodiments, the three-dimensional directions can also be replaced by cylindrical axis coordinates.

[0064] In each first grid space, cluster analysis is performed on the point cloud to detect obstacles in the surrounding environment, and the covariance matrix of each cluster is calculated to solve for the corresponding eigenvalues ​​and eigenvectors. Then, the coordinate points are projected onto the direction vectors to find the maximum and minimum values ​​of the x, y, and z components in each direction, and the minimum bounding box of the obstacles, i.e., the "first bounding box", is constructed.

[0065] Optionally, the second external bounding box includes a first sub-bounding box of the coordinate system corresponding to the multi-boom equipment and a second sub-bounding box of the coordinate system corresponding to the first boom. The pump truck boom is divided into two levels of bounding spheres. The first level bounding sphere is a spherical space centered on the boom center of the multi-boom equipment, with the radius being the furthest distance from the center to the boom. This is the "first sub-bounding box." The second level bounding sphere is the spherical space formed by a single boom section, which is the "second sub-bounding box."

[0066] Please combine Figure 1 and Figure 2 Optionally, in S3, during the first collision detection based on the first and second bounding boxes, the method includes:

[0067] The first and second outer bounding boxes perform the first collision detection in the coordinate system of the multi-arm equipment.

[0068] Specifically, the coordinate system of the first bounding box of the obstacle and the coordinate system of the pump truck boom are both unified to the coordinate system of the multi-boom equipment, so that the first and second bounding boxes can perform the first collision detection in the coordinate system of the multi-boom equipment. During detection, the first sub-bounding box is used as the maximum detection space for collision detection. In the coordinate system of the multi-boom equipment, each detected first bounding box is treated as an independent obstacle sensing object, and the first collision detection is performed with each second sub-bounding box to determine whether there is an obstacle and boom that will collide.

[0069] Optionally, in S3, the method for performing the first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom device includes:

[0070] Determine whether the first outer bounding box and the first sub-bounding box intersect;

[0071] If the first outer bounding box and the first sub-bounding box do not intersect, the result of the first collision detection indicates that there is no first obstacle and the first boom that collided;

[0072] If the first outer bounding box intersects with the first sub-bounding box, the distance between the first outer bounding box and the second sub-bounding box is used to determine whether the conditions for a collision are met.

[0073] If the conditions for a collision are met, the first obstacle and the first boom that collided are identified, and the result of the first collision detection indicates that there is a first obstacle and the first boom that collided.

[0074] If the conditions for a collision are not met, the result of the first collision detection indicates that there is no first obstacle and first boom that will collide.

[0075] Specifically, the minimum distance between the first outer bounding box and the second sub-bounding box can be taken, and it can be determined whether the minimum value is within the collision range. If the minimum value is within the collision range, the conditions for a collision are met, and there is a first obstacle and a first boom that will collide. Otherwise, the conditions for a collision are not met, and there is no first obstacle and a first boom that will collide.

[0076] In cases where the first collision detection indicates the presence of a first obstacle and a first boom that could collide, a second collision detection is performed to conduct a more precise collision avoidance detection and more accurately determine whether a collision risk actually exists.

[0077] Please combine Figure 1 and Figure 2 Optionally, in S4, based on the second point cloud data, the second collision detection is performed in the coordinate system corresponding to the first boom, and the method includes:

[0078] Voxel subdivision is performed on the 3D point cloud corresponding to the second point cloud data;

[0079] Based on the results of the voxel subdivision, multiple third bounding boxes corresponding to the first obstacle are established;

[0080] The second collision detection is performed based on the coordinates of the third outer bounding box in the coordinate system corresponding to the first arm.

[0081] In this embodiment, the 3D point cloud corresponding to the second point cloud data is divided into voxel subdivisions along the x, y, and z axes, with a subdivision interval of a second preset distance. This second preset distance is less than a first preset distance; in this embodiment, the second preset distance is 0.05-0.3 meters, preferably 0.1 meters, thus forming a W*M*N second grid space (W = X / s, M = Y / s, N = Z / s, where X, Y, and Z represent the side lengths of the 3D point cloud corresponding to the second point cloud data along the x, y, and z axes). In other embodiments, the three-dimensional directions can be replaced with cylindrical axis coordinates.

[0082] Subsequently, within each second grid space, cluster analysis is performed on the point cloud to detect obstacles within the second grid space. The covariance matrix of each cluster is calculated, and the corresponding eigenvalues ​​and eigenvectors are solved. Then, the coordinate points are projected onto the direction vectors, and the maximum and minimum values ​​of the x, y, and z components in each direction are found to construct multiple minimum bounding boxes (MUBs) for the first obstacle, also known as "third bounding boxes." In this way, when irregularly shaped first obstacles exist, the first obstacle is represented using multiple MUBs, further improving accuracy compared to the first collision detection method.

[0083] Next, the coordinate system of the third external enclosure box is transferred to the coordinate system corresponding to the first boom for the second collision detection. Compared with the unification to the vehicle body coordinate system in the coarse-level detection, this reduces intermediate coordinate transformation steps and reduces errors.

[0084] Please combine Figure 1 and Figure 2 During the second collision detection, the minimum distance between the third outer bounding box and the second sub-bounding box can be taken to determine if the minimum value is within the alarm range. If the minimum value is within the alarm range, it indicates that the first obstacle and the first boom may collide, generating a collision avoidance alarm. Conversely, it indicates that the first obstacle and the first boom will not collide, and no collision avoidance alarm is generated.

[0085] It should be noted that the first and second collision detections can be performed in real time during boom movement, or they can be performed after the boom has stopped moving. Alternatively, the first collision detection can be performed in real time during boom movement, while the second collision detection can be performed after the boom has stopped moving, in order to simultaneously ensure the efficiency and safety of boom control.

[0086] Optionally, after S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes the first bounding box of the obstacles, the method includes:

[0087] Obstacle avoidance path planning is performed based on the first outer bounding box.

[0088] Specifically, when the first collision detection indicates the presence of a first obstacle and the first boom colliding, obstacle avoidance path planning is performed based on the first bounding box. Specifically, each detected first bounding box is treated as an independent obstacle sensing object and participates in the real-time dynamic obstacle avoidance path planning of the boom to form an optimal obstacle avoidance route for obstacle avoidance control. Thus, using the first bounding box as the sensing object allows for more thorough obstacle avoidance.

[0089] This application utilizes multi-joint boom active collision avoidance technology to achieve collision detection, alarm, and obstacle avoidance functions between the boom and unknown obstacles in the surrounding space. The entire process requires no manual intervention, improving the intelligence level of the equipment and meeting the requirements of digital construction.

[0090] Furthermore, to better achieve automatic control of the boom, ensuring that the boom's pose meets the requirements of lidar acquisition area covering the entire boom space and achieving optimal construction conditions, a boom motion model can be trained based on the Deep Deterministic Policy Gradient (DDPG) algorithm. During training, a reward and penalty mechanism associated with motion parameters such as boom posture, position, and number of action segments can be used to recommend boom actions to meet the needs of reasonable boom movement and / or achieving the desired state.

[0091] Specifically, the method for training the boom motion model includes: creating an Actor network and a Critic network; interacting with the environment through the Actor network to determine the second state reached by the boom in the first state by taking the first action and the reward value obtained. The preset reward and punishment strategy used for the reward value includes a reward and punishment mechanism associated with at least one of the following: the target posture of the boom, the target position of the boom end effector, and the number of boom action segments; storing the first state, the first action, the reward value, and the second state in a training database; and training the Actor network and the Critic network to obtain the trained boom motion model.

[0092] Optionally, the preset reward and penalty strategy includes a reward and penalty mechanism associated with the target posture of the boom, the target position of the boom end effector, and the number of boom action segments. This mechanism interacts with the environment through an Actor network to determine the second state reached by the boom in the first state after taking the first action, and the resulting reward value, including:

[0093] By interacting with the environment through an Actor network, the second state reached by the boom in the first state after taking the first action is determined;

[0094] The end position of the boom is determined based on the second state, and the distance between the end position of the boom and the target end position of the boom is obtained; the attitude of the boom is determined based on the second state; the number of boom action segments is determined based on the first action.

[0095] The reward value is determined based on a preset reward and punishment strategy, which is as follows:

[0096] reward = r + r1 + r2 + r3

[0097] Wherein, reward is the reward value; r is the initial value, which is a negative value of the distance value; r1 is the reward / penalty value associated with the boom end position. If the distance value is within the preset range, r1 is a positive value; if the distance value is not within the preset range, r1 is a negative value; r2 is the reward / penalty value associated with the boom posture. If the distance value is within the preset range and the boom posture is similar to the boom target posture, r2 is a positive value; if the distance value is within the preset range and the boom posture is not similar to the boom target posture, r2 is a negative value; r3 is the reward / penalty value associated with the number of boom action segments. When the distance value is within the preset range and the boom posture is similar to the boom target posture, r3 is determined according to the number of boom action segments and is a positive value.

[0098] First, the entire movable area of ​​the boom is used as the training region for unified training, resulting in a trained boom motion model. The training results are then checked to confirm if they meet the accuracy requirements, such as a measurement accuracy of 0.1 to 5 meters. If the training results do not meet expectations, the algorithm parameters are adjusted, and the training / test set is redistributed, to continue training for the next iteration. This significantly reduces the time spent on the initial "exploration" process during network training. Simultaneously, considering that the training process does not consume many system resources, the training region can be divided into multiple smaller regions based on their distance from the slewing center. Multiple smaller regions are trained synchronously during the training process. After training, a partitioning selection strategy can achieve the same effect as unified training, greatly reducing the training set size required for algorithm convergence and shortening the time required to complete a single training iteration.

[0099] During training, to ensure that the posture solved by the network conforms to the joint limits and other conditions required by the actual working conditions of the boom, and to prioritize the "arched" posture, the reward and penalty function of the optimization algorithm is optimized to minimize the number of moving booms, recommending an approximate "arched" target point posture. Simultaneously, to reduce large boom movements, corresponding boom locking strategies and constraints limiting the boom's range of motion are set. Furthermore, corresponding penalty / incentive strategies are designed for different boom postures to guide the network training towards the expected results, ultimately solving for the recommended posture of the equipment at the target location. In actual implementation, the recommended posture can also be an arched posture, an M-shaped posture, an L-shaped posture, or a prone posture. The arched, M-shaped, and L-shaped postures are for boom operations on high-rise buildings, while the prone posture is for boom operations on underground structures, increasing the applicability of the collision avoidance detection method of this invention. In practical operation, the arched posture is mostly used for boom operations on high-rise buildings; therefore, in this embodiment, the arched posture is preferred as the boom operation mode.

[0100] The model described above can output recommended actions to bring the boom to a recommended posture. Then, based on the strategy, the boom's actions are adjusted to ensure its pose covers the entire boom space for LiDAR data acquisition, meeting the requirements for point cloud data collection. Furthermore, when there is a risk of collision with obstacles, recommended actions to reach the recommended posture can be re-output based on the boom's current pose to ensure the boom's pose reaches the optimal state for construction, while still maintaining LiDAR coverage of the entire boom space during subsequent collision detection.

[0101] As described above, this invention discloses a boom collision avoidance detection method, comprising: acquiring first point cloud data of the environment surrounding a multi-boom device; detecting obstacles in the surrounding environment based on the first point cloud data and establishing a first bounding box for the obstacles; performing a first collision detection based on the first bounding box and a second bounding box for the boom in the multi-boom device; if the result of the first collision detection indicates the existence of a first obstacle and a first boom colliding, performing a second collision detection based on the point cloud data corresponding to the first obstacle in the first point cloud data, in the coordinate system corresponding to the first boom; and generating a collision avoidance alarm information if the result of the second collision detection indicates a collision between the first obstacle and the first boom. This invention can automatically perform collision detection, and by performing a first collision detection based on the point cloud data of the surrounding environment, and then performing a second collision detection based on the point cloud data of the obstacle colliding, the detection results are more accurate.

[0102] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described above.

[0103] This application also provides a computer-readable storage medium storing a computer program that performs the steps of any of the methods described above.

[0104] Please combine Figure 3 and Figure 4 This application also provides a multi-arm device 700, including a lidar 702 and a controller 706. The lidar 702 is used to collect data to obtain the first point cloud data, and the controller 706 executes a computer program to implement the steps of the method described in the above embodiments.

[0105] The multi-arm boom 700 may further include a camera 701, a switch 703, an image processor 704, and an alarm device 705. The camera 701 and the lidar 702 are mounted on the multi-arm boom 707, while the switch 703, image processor 704, and alarm device 705 are mounted in the operator's cab 708. The image processor 704 and controller 706 include multiple circuit function modules. The alarm device 705 may be a horn, a buzzer, or a display screen; this invention is not limited to these.

[0106] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for detecting collision avoidance of a boom, characterized in that, Applied to multi-arm equipment, the method includes: S1 acquires the first point cloud data of the environment surrounding the multi-arm device; S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes a first bounding box for the obstacles; S3 performs a first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom device; S4 If the result of the first collision detection indicates that there is a first obstacle and the first boom colliding, multiple third bounding boxes corresponding to the first obstacle are established based on the second point cloud data. In the coordinate system corresponding to the first boom, a second collision detection is performed according to the distance between the third bounding box and the second sub-bounding box corresponding to the first boom. The obstacles in the surrounding environment include the first obstacle, the multi-boom device includes the first boom, and the second point cloud data is the point cloud data corresponding to the first obstacle in the first point cloud data. S5 generates a collision avoidance alarm if the result of the second collision detection indicates that the first obstacle has collided with the first boom.

2. The method as described in claim 1, characterized in that, In S4, based on the second point cloud data, a second collision detection is performed in the coordinate system corresponding to the first boom. The method includes: Voxel subdivision is performed on the 3D point cloud corresponding to the second point cloud data; Based on the results of the voxel subdivision, multiple third bounding boxes corresponding to the first obstacle are established; The second collision detection is performed based on the coordinates of the third external bounding box in the coordinate system corresponding to the first boom.

3. The method as described in claim 1, characterized in that, In S3, during the first collision detection based on the first bounding box and the second bounding box, the method includes: The first external bounding box and the second external bounding box perform the first collision detection in the coordinate system of the multi-arm equipment.

4. The method as described in claim 1, characterized in that, After S2 detects obstacles in the surrounding environment based on the first point cloud data and establishes a first bounding box for the obstacles, the method includes: Obstacle avoidance path planning is performed based on the first external bounding box.

5. The method as described in claim 1, characterized in that, The second external bounding box includes a first sub-bounding box of the coordinate system corresponding to the multi-arm device and a second sub-bounding box of the coordinate system corresponding to the first arm; In S3, during the first collision detection based on the first external enclosure box and the second external enclosure box of the boom in the multi-boom device, the method includes: Determine whether the first outer bounding box intersects with the first sub-bounding box; If the first outer bounding box and the first sub-bounding box do not intersect, the result of the first collision detection indicates that there is no first obstacle and the first boom that collided; If the first outer bounding box intersects with the first sub-bounding box, the distance between the first outer bounding box and the second sub-bounding box is used to determine whether the conditions for a collision are met. If the conditions for a collision are met, the first obstacle and the first boom that collided are identified, and the result of the first collision detection indicates that there is a first obstacle and the first boom that collided. If the conditions for a collision are not met, the result of the first collision detection indicates that there is no first obstacle and first boom that will collide.

6. The method as described in claim 1, characterized in that, In acquiring the first point cloud data of the environment surrounding the multi-arm equipment in S1, the method includes: When the multi-arm device rotates to the left, the point cloud data corresponding to the left side of the arm obtained from the lidar data is used as the first point cloud data; When the multi-arm device rotates to the right, the point cloud data corresponding to the right side of the arm obtained from the lidar data is used as the first point cloud data; When the multi-arm device undergoes a luffing action, the point cloud data corresponding to a preset range around the central axis of the arm obtained from the lidar data is used as the first point cloud data.

7. The method as described in claim 1, characterized in that, In step S2, which detects obstacles in the surrounding environment based on the first point cloud data and establishes a first bounding box for the obstacles, the method includes: The three-dimensional point cloud corresponding to the first point cloud data is divided into voxel subdivisions to establish multiple first grid spaces; Cluster analysis is performed on the point cloud within the first grid space to detect obstacles in the surrounding environment and determine the first bounding box of the obstacles.

8. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 7.

9. A multi-arm scaffolding device, characterized in that, The system includes a lidar and a controller, wherein the lidar is used to collect data to obtain the first point cloud data, and the controller executes a computer program to implement the method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that, The readable storage medium stores a computer program that performs the method of any one of claims 1 to 7.

Images