Autonomous withdrawing method and device of robot, electronic equipment and storage medium

Abstract

The invention provides an autonomous withdrawing method and device of a robot, electronic equipment and a storage medium, and relates to the technical field of robots. The autonomous withdrawing method of the robot comprises the following steps: in response to a received starting instruction of a target inspection task, obtaining position information of an initial safety point; starting to execute a target inspection task based on the position information of the initial safety point; in the process of executing the target inspection task, in response to the situation that the change of the motion state of the robot at any moment relative to the previous moment is greater than a preset threshold value, obtaining and storing motion state update data at the moment; in response to the fact that the autonomous retracement condition is met, target retracement data at the current moment are obtained according to stored motion state updating data at multiple moments; and according to the retracement encoder increment information, executing retracement operation so as to reach the target position. According to the invention, the robot can be safely and accurately withdrawn in an emergency scene.

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and in particular to a robot autonomous withdrawal method, device, electronic equipment and storage medium. Background Art

[0002] In the current field of robotics, a common problem is that most robots lack a retracement mechanism. When the signal is interrupted, the robot often loses control and cannot continue to perform the task, and can only stay where it is, resulting in the inability to transmit the collected data back in time, which seriously affects the robot's operating efficiency and reliability.

[0003] In order to deal with the control problems caused by signal interruption, some simple retraction methods have emerged, but these methods are not accurate and cannot cope with emergency scenarios, resulting in large errors in the generated robot return route. Not only is the return route long, but it also takes a long time, which cannot meet the needs of actual application scenarios. Especially in emergency scenarios, when the environment changes drastically (such as fire or collapse at the scene), the robot cannot obtain an accurate and safe return route.

[0004] Therefore, how to make the robot retreat safely and accurately in emergency scenarios is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of the above problems existing in the prior art, the present invention provides a method, device, electronic device and storage medium for autonomous withdrawal of a robot, so as to enable the robot to withdraw safely and accurately in an emergency scenario.

[0006] The present invention provides an autonomous withdrawal method of a robot, comprising the following steps.

[0007] In response to receiving a start instruction for a target inspection task, the position information of an initial safety point is obtained; based on the position information of the initial safety point, the target inspection task is started; in the process of executing the target inspection task, in response to the change of the robot's motion state at any moment relative to the previous moment being greater than a preset threshold, the motion state update data at the moment is obtained and stored; wherein, the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; in response to meeting the autonomous withdrawal condition, the target withdrawal data at the current moment is obtained according to the motion state update data stored at multiple moments; wherein, the target withdrawal data includes the target position and the withdrawal encoder increment information, and the target position satisfies at least one or more of the following conditions: no collapse, no fire and no obstruction by obstacles; according to the withdrawal encoder increment information, a withdrawal operation is performed to reach the target position, and the next target withdrawal data is continuously obtained according to the stored multiple motion state update data until the initial safety point is reached.

[0008] According to a robot autonomous withdrawal method provided by the present invention, the obtaining of the position information of the initial safety point includes: using a laser radar to obtain point cloud information of the current position, and using a camera to obtain visual information of the current position; using a preset fusion algorithm, based on the point cloud information and the visual information, obtaining the position information of the initial safety point.

[0009] According to an autonomous withdrawal method of a robot provided by the present invention, the obtaining and storing of the motion state update data at the moment includes: obtaining the encoder increment information at the moment according to the instantaneous value of the encoder at the moment and the instantaneous value of the encoder at the previous moment; using an information fusion algorithm to obtain the initial values ​​of the position and posture of the robot at the moment according to the value of the inertial measurement unit IMU at the moment and the encoder increment information at the moment; using an iterative closest point ICP algorithm to correct the initial values ​​of the position and posture at the moment according to the point cloud information of the position at the moment obtained by a laser radar and the visual information of the position at the moment obtained by a camera device, and storing the corrected position, posture and the encoder increment information as the motion state update data at the moment.

[0010] According to a robot autonomous withdrawal method provided by the present invention, the response to satisfying the autonomous withdrawal condition includes: confirming that the autonomous withdrawal condition is satisfied when the communication signal of the robot is interrupted; or confirming that the autonomous withdrawal condition is satisfied when the ambient temperature obtained by using an ambient temperature detection device is higher than a preset temperature threshold.

[0011] According to a robot autonomous retraction method provided by the present invention, the target retraction data at the current moment is obtained based on the motion state update data at multiple moments that have been stored, including: selecting the motion state update data closest to the current moment from the motion state update data at the multiple moments as candidate motion state update data; determining whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by using a laser radar and the visual information obtained by using a camera device; wherein the retractable point is a position point that at least satisfies one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; when the position included in the candidate motion state update data is a retractable point, a preset inverse motion algorithm is used to obtain candidate retraction data corresponding to the candidate motion state update data based on the candidate motion state update data; wherein the candidate retraction data includes a first candidate retraction position and a candidate retraction encoder increment; according to the point cloud information obtained by the laser radar and the visual information obtained by using a camera device, the current position of the robot is obtained; according to the candidate retraction encoder increment, an ICP algorithm is used, The current position of the robot is iteratively adjusted until a position adjustment termination condition is reached to obtain a second candidate withdrawal position; wherein the position adjustment termination condition includes: the difference between the first candidate withdrawal position and the second candidate withdrawal position is less than a preset error threshold, or a preset number of adjustments is reached; in response to the adjustment result that the difference between the first candidate withdrawal position and the second candidate withdrawal position is greater than the preset error threshold, the next motion state update data closest to the current moment is selected from the motion state update data stored at the multiple moments as the candidate motion state update data, and the step of determining whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by using the laser radar and the visual information obtained by using the camera device and subsequent steps are repeatedly executed until the distance between the first candidate withdrawal position and the second candidate withdrawal position obtained by recalculation is less than the preset error threshold; in response to the adjustment result that the difference between the first candidate withdrawal position and the second candidate withdrawal position is less than the preset error threshold, the target withdrawal data is obtained according to the candidate withdrawal data.

[0012] According to a robot autonomous withdrawal method provided by the present invention, the method also includes: when the position included in the candidate motion state update data is a non-retreatable point, determining a detour position using point cloud information obtained according to the laser radar and visual information obtained by the camera device; wherein the detour position is located on a path, the path is perpendicular to the path from the position included in the candidate motion state update data to the current position of the robot, the path passes through the position included in the candidate motion state update data, and the detour position satisfies at least one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; according to the detour position, updating the position in the motion state update data.

[0013] The present invention also provides an autonomous retraction device for a robot, comprising the following modules: A first acquisition module is used to obtain the position information of the initial safety point in response to receiving a start instruction of the target inspection task; an execution module is used to start executing the target inspection task based on the position information of the initial safety point; a second acquisition module is used to obtain and store the motion state update data at any moment in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold during the execution of the target inspection task; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; a third acquisition module is used to obtain the target withdrawal data at the current moment according to the motion state update data of multiple moments stored in response to meeting the autonomous withdrawal condition; wherein the target withdrawal data includes the target position and the withdrawal encoder increment information, and the target position at least meets one or more of the following conditions: no collapse, no fire and no obstacle; a withdrawal module is used to perform a withdrawal operation according to the withdrawal encoder increment information to reach the target position, and continue to obtain the next target withdrawal data according to the stored multiple motion state update data until the initial safety point is reached.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the autonomous withdrawal method of the robot as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the autonomous withdrawal method of the robot as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the autonomous withdrawal method of the robot as described above is implemented.

[0017] The autonomous withdrawal method, device, electronic device and storage medium of the robot provided by the present invention, in the process of executing the target inspection task, when the change of the robot's motion state at any moment relative to the previous moment is greater than a preset threshold, the motion state update data at the moment is obtained and stored, so as to provide the robot with return path information when it needs to withdraw; when the autonomous withdrawal conditions are met, the target withdrawal data at the current moment is obtained according to the motion state update data stored at multiple moments, and the target withdrawal data includes the target position and the withdrawal encoder increment information, and the target position at least meets one or more of the following conditions: no collapse, no fire and no obstacles. Therefore, when the robot encounters an emergency (for example, a fire), it can accurately and safely carry the collected data information and withdraw to a safe place according to the target withdrawal data, which effectively improves the robot's ability to cope with emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a schematic flow chart of the autonomous withdrawal method of the robot provided by the present invention.

[0020] Figure 2 It is a flow chart of the method for obtaining motion status update data provided by the present invention.

[0021] Figure 3 It is a flow chart of the method for obtaining target withdrawal data provided by the present invention.

[0022] Figure 4 It is a schematic structural diagram of the autonomous retraction device of the robot provided by the present invention.

[0023] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Combine the following Figure 1-Figure 3 The autonomous withdrawal method of the robot of the present invention is described.

[0026] Figure 1 FIG. 1 is a flow chart of the autonomous retraction method of the robot provided by the present invention, and the method is applied to the robot. Figure 1 As shown, the method includes the following: Step 101: In response to receiving a start instruction of a target inspection task, obtaining location information of an initial safety point.

[0027] As an example only, the target inspection task may be to conduct autonomous inspections in a substation to check the operating status and parameters of equipment such as transformers, distribution cabinets, and relay protection devices.

[0028] The start instruction of the target inspection task may include but is not limited to: power on, a button corresponding to the target inspection task is pressed, etc.

[0029] The robot is equipped with a variety of sensors, including encoders, inertial measurement units (IMUs), lidars, cameras (e.g., RGB cameras), etc.

[0030] In some embodiments, the location information of the initial safety point can be obtained by using the following methods: using a laser radar to obtain point cloud information of the current location, and using a camera to obtain visual information of the current location; using a preset fusion algorithm (for example, a Kalman filter algorithm, etc.) to obtain the location information of the initial safety point based on the point cloud information and visual information.

[0031] Step 102: Based on the location information of the initial safety point, start executing the target inspection task.

[0032] During the specific implementation process, the robot can use lidar and camera equipment to perform SLAM mapping and navigation, start autonomous cruising, and begin to perform target inspection tasks according to the work requirements of the target inspection task, starting from the initial safety point.

[0033] Step 103: During the execution of the target inspection task, in response to a change in the robot's motion state at any moment relative to the previous moment being greater than a preset threshold, acquiring and storing motion state update data at that moment.

[0034] The motion state includes the position or posture of the robot.

[0035] The motion state update data includes the encoder increment information of the robot at this moment relative to the previous moment and the position and posture at this moment.

[0036] In the specific implementation process, the updated data of the robot's motion state at that moment can be obtained in a variety of ways, which are not limited to the description of this specification.

[0037] For an embodiment of obtaining the updated data of the robot's motion state at this moment, see Figure 2 The relevant content in will not be repeated here.

[0038] Step 104 : In response to satisfying the autonomous withdrawal condition, update the data according to the motion states stored at multiple moments to obtain the target withdrawal data at the current moment.

[0039] During the robot's inspection process, if there are unexpected situations such as signal shielding, co-frequency crosstalk, over-distance communication interruption, and rapid temperature rise caused by explosion and combustion, the operator's control over the robot will fail. In order to protect the robot and the original data or samples it collects, the robot can use ambient temperature detection equipment (for example, infrared cameras) to obtain and report the surrounding ambient temperature in real time during the inspection process.

[0040] In the specific implementation process, when the communication signal of the robot is interrupted, it is confirmed that the autonomous withdrawal condition is met; or, when the ambient temperature obtained by the ambient temperature detection device is higher than the preset temperature threshold, it is confirmed that the autonomous withdrawal condition is met.

[0041] The target retraction data includes the target position and the incremental information of the retraction encoder. The target position at least satisfies one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles.

[0042] The target position is the retraction position that the robot needs to reach at the next moment. The retraction encoder increment information is the encoder increment information of the robot from the current position to the target position.

[0043] For more information about updating data based on the motion states stored at multiple times and obtaining the target retracement data at the current moment, see Figure 3 The relevant content in will not be repeated here.

[0044] Step 105 , perform a retraction operation according to the incremental information of the retraction encoder to reach the target position, and continue to obtain the next target retraction data according to the stored multiple motion state update data until the initial safety point is reached.

[0045] During the specific implementation process, the robot can calculate the path required to withdraw to the target position based on the incremental information of the withdrawal encoder combined with the current position and posture information; according to the calculated withdrawal path, adjust its motion control parameters, such as speed, direction, etc., to start the withdrawal operation.

[0046] After reaching the target position, the robot repeats step 104 to step 105 until reaching the initial safety point.

[0047] Figure 2 is a flow chart of a method for obtaining motion status update data provided by the present invention, such as Figure 2 As shown, the method includes the following: Step 201: Obtain encoder increment information at any moment according to the encoder instantaneous value at any moment and the encoder instantaneous value at the previous moment.

[0048] In the specific implementation process, the robot can record the instantaneous values ​​E of the left front, left rear, right front and right rear encoders of the encoder. k,i (k=1,…,4, respectively the instantaneous values ​​of the left front, left rear, right front, and right rear encoders, i=1,…,N, N represents the number of recorded status information), as well as IMU information.

[0049] By comparing the instantaneous value of the encoder at this moment with the instantaneous value of the encoder recorded at the previous moment, the encoder increment information ∆E at this moment is calculated. k,i (k=1,…,4, respectively the left front, left rear, right front and right rear encoder increment values, i=1,…,N, N represents the number of times the motion state data is updated).

[0050] Step 202: Using an information fusion algorithm, according to the value of the inertial measurement unit IMU and the encoder increment information at that moment, the initial values ​​of the position and posture of the robot at that moment are obtained.

[0051] In the specific implementation process, the information fusion algorithm can be used to fuse the value of the inertial measurement unit IMU at that moment and the encoder increment information at that moment to obtain the robot's position and posture p at that moment. i (including the robot's posture information and position information, i=1,…,N, N represents the number of times the recorded motion state data is updated).

[0052] Step 203: Using the ICP algorithm, based on the point cloud information of the position at that moment obtained by the laser radar and the visual information of the position at that moment obtained by the camera, the initial values ​​of the position and posture at that moment are corrected, and the corrected position, posture and encoder increment information are stored as the motion state update data at that moment.

[0053] As an example only, the robot's position p at this moment obtained in the above steps is i The ICP algorithm can be used to correct p based on the point cloud information of the robot's location at that moment obtained by the laser radar, the visual information of the robot's location at that moment obtained by the camera, and the IMU information. i P i (including the robot's posture information and position information, i=1,…,N, N represents the number of times the recorded motion state data is updated).

[0054] (1) in, is the robot's position information in the x direction, is the robot's position information in the y direction, is the robot's position information in the z direction, is the pitch angle of the robot, is the yaw angle of the robot.

[0055] In a specific implementation process, at multiple moments in the inspection process of the robot, the corrected position and encoder increment information obtained by the above steps are stored as the motion state update data at that moment, thereby obtaining multiple state update data.

[0056] Figure 3 is a flow chart of a method for obtaining target retracement data provided by the present invention, such as Figure 3 As shown, the method includes the following: Step 301: Select the motion state update data closest to the current moment from the motion state update data at multiple moments as candidate motion state update data.

[0057] As an example only, the saved motion state update data P can be popped out one by one from the back i , ∆E k,i Information, at each step of the retracement, select the motion state update data closest to the current moment as the candidate motion state update data.

[0058] Step 302: Determine whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by the laser radar and the visual information obtained by the camera.

[0059] The retractable point is a position point that meets at least one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles. These conditions ensure that the robot can safely retract to this position.

[0060] In the specific implementation process, for the location included in the candidate motion state update data, the topography and obstacle distribution of the location point can be analyzed using the lidar point cloud information. If the point cloud data shows that the location point has signs of collapse (such as uneven terrain, deep pits, etc.), it is judged that the location point does not meet the "non-collapse" condition; if the point cloud data shows that there are obstacles around the location point, and the distance between the obstacle and the location point is less than the safety distance threshold, it is judged that the location point does not meet the "non-obstructed by obstacles" condition.

[0061] In the specific implementation process, for the position included in the candidate motion state update data, visual information analysis can be performed using a camera device: for the position point, the state of the position point is analyzed using the visual information of the camera device; if features such as flames or smoke are identified in the visual information, indicating that the position point or the nearby area may be on fire, then it is determined that the position point does not meet the "not on fire" condition.

[0062] If the position included in the candidate motion state update data satisfies all conditions at the same time (including but not limited to not collapsing, not catching fire, not being blocked by obstacles, etc.), the position point is determined to be a retractable point; otherwise, the position point is a non-retractable point.

[0063] In some embodiments, when the position included in the candidate motion state update data is a non-retreat point, the detour position is determined using point cloud information obtained based on a laser radar and visual information obtained using a camera device; wherein the detour position is located on a path that is perpendicular to the path from the position included in the candidate motion state update data to the current position of the robot, the path passes through the position included in the candidate motion state update data, and the detour position satisfies at least one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; based on the detour position, the position in the motion state update data is updated.

[0064] As an example, a robot can use LiDAR to scan the surrounding environment and obtain accurate 3D point cloud data, which contains detailed information such as terrain, obstacles, roads, etc. It can also use cameras to capture images of the surrounding environment and perform image processing to identify key information such as flames, smoke, and obstacle types. Combining point cloud information and visual information, the robot can internally build a 3D model or 2D map of the surrounding environment to provide a basis for path planning.

[0065] The robot can use a path planning algorithm (such as the A* algorithm, Dijkstra algorithm, etc.) to plan a new path perpendicular to the path from the non-retractable point to the current position based on the above-mentioned three-dimensional model or two-dimensional map. The path passes through the non-retractable point and is used to find a detour position. On the new path, the robot can select a position that meets the conditions of not collapsing, not on fire, and not blocked by obstacles as the detour position.

[0066] Step 303: When the position included in the candidate motion state update data is a retractable point, a preset inverse motion algorithm is used to obtain candidate retractable data corresponding to the candidate motion state update data according to the candidate motion state update data.

[0067] The candidate pullback data includes a first candidate pullback position and a candidate pullback encoder increment.

[0068] As an example only, the data P is updated for the candidate motion state i , ∆E k,i , the candidate retracement data is calculated using the preset inverse motion algorithm, which includes the first candidate retracement position RP i , and the candidate pullback encoder increment ∆RE k,i .

[0069] Step 304: Obtain the current position of the robot based on the point cloud information obtained by the laser radar and the visual information obtained by the camera.

[0070] Step 305: Using the ICP algorithm, according to the candidate retraction encoder increment, the current position of the robot is iteratively adjusted until the position adjustment termination condition is reached, thereby obtaining a second candidate retraction position.

[0071] The position adjustment termination conditions include: the difference between the first candidate retracement position and the second candidate retracement position is less than a preset error threshold, or the preset adjustment times are reached.

[0072] As an example only, the candidate pullback encoder increment ∆RE can be calculated based on k,i , Use the ICP algorithm to adjust the robot's current position, and use the adjusted current position as the second candidate retracement position , until RP i -P i <∆e, ∆e is the preset error threshold: (2) in, is the position information of the second candidate retracement position in the x direction, is the position information of the second candidate retracement position in the y direction, The position information of the second candidate retracement position in the z direction; is the position information of the first candidate retracement position in the x direction, is the position information of the first candidate retracement position in the y direction, The position information of the first candidate retracement position in the z direction; is the pitch angle at the current moment, is the yaw angle at the current moment; is the pitch angle of the candidate retracement data record, The yaw angle recorded for the candidate pullback data.

[0073] In the specific implementation process, when the difference between the first candidate retraction position and the second candidate retraction position is greater than the preset error threshold, the next motion state update data closest to the current moment is selected from the motion state update data of multiple moments stored as the candidate motion state update data, and the step of determining whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by using the lidar and the visual information obtained by using the camera equipment and subsequent steps are repeatedly executed until the distance between the recalculated first candidate retraction position and the second candidate retraction position is less than the preset error threshold.

[0074] Step 306: In response to the adjustment result that the difference between the first candidate retracement position and the second candidate retracement position is less than a preset error threshold, obtain target retracement data according to the candidate retracement data.

[0075] In a specific implementation process, when the adjustment result is that the difference between the first candidate retracement position and the second candidate retracement position is less than a preset error threshold, the candidate retracement data may be used as the target retracement data.

[0076] The autonomous withdrawal device of the robot provided by the present invention is described below. The autonomous withdrawal device of the robot described below and the autonomous withdrawal method of the robot described above can be referred to each other.

[0077] Figure 4 Schematic diagram of the structure of the autonomous retraction device of the robot provided by the present invention. Figure 4 As shown, the device 400 includes the following modules.

[0078] The first acquisition module 410 is used to acquire the position information of the initial safety point in response to receiving the start instruction of the target inspection task.

[0079] The execution module 420 is used to start executing the target inspection task based on the location information of the initial safety point.

[0080] The second acquisition module 430 is used to acquire and store the motion state update data at any moment relative to the previous moment in response to a change of the motion state of the robot at any moment being greater than a preset threshold during the execution of the target inspection task; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder incremental information of the robot at the previous moment and the position and posture of the robot at the moment.

[0081] The third acquisition module 440 is used to respond to the autonomous retraction condition being met, update the data according to the motion state at multiple stored moments, and obtain the target retraction data at the current moment; wherein the target retraction data includes the target position and the retraction encoder increment information, and the target position satisfies at least one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles.

[0082] The retraction module 450 is used to perform a retraction operation according to the retraction encoder increment information to reach the target position, and continue to obtain the next target retraction data according to the stored multiple motion state update data until reaching the initial safety point.

[0083] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the autonomous withdrawal method of the robot, the method comprising: in response to receiving the start instruction of the target inspection task, obtaining the position information of the initial safety point; based on the position information of the initial safety point, starting to execute the target inspection task; in the process of executing the target inspection task, in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold, obtaining and storing the motion state update data at the moment; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; in response to satisfying the autonomous withdrawal condition, obtaining the target withdrawal data at the current moment according to the motion state update data stored at multiple moments; Among them, the target retraction data includes the target position and the retraction encoder incremental information, and the target position satisfies at least one or more of the following conditions: not collapsed, not on fire and not blocked by obstacles; according to the retraction encoder incremental information, a retraction operation is performed to reach the target position, and the next target retraction data is obtained according to the stored multiple motion state update data until the initial safety point is reached.

[0084] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0085] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the autonomous withdrawal method of the robot provided by the above methods, the method comprising: in response to receiving a start instruction of a target inspection task, obtaining the position information of an initial safety point; based on the position information of the initial safety point, starting to execute the target inspection task; in the process of executing the target inspection task, in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold, obtaining and storing the motion state update data at the moment; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; in response to satisfying the autonomous withdrawal condition, obtaining the target withdrawal data at the current moment according to the motion state update data stored at multiple moments; Among them, the target retraction data includes the target position and the retraction encoder incremental information, and the target position satisfies at least one or more of the following conditions: not collapsed, not on fire and not blocked by obstacles; according to the retraction encoder incremental information, a retraction operation is performed to reach the target position, and the next target retraction data is obtained according to the stored multiple motion state update data until the initial safety point is reached.

[0086] On the other hand, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the autonomous withdrawal method of the robot provided by the above-mentioned methods, the method comprising: in response to receiving a start instruction of a target inspection task, obtaining the position information of an initial safety point; based on the position information of the initial safety point, starting to execute the target inspection task; in the process of executing the target inspection task, in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold, obtaining and storing the motion state update data at the moment; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; in response to satisfying the autonomous withdrawal condition, obtaining the target withdrawal data at the current moment according to the motion state update data stored at multiple moments; Among them, the target retraction data includes the target position and the retraction encoder incremental information, and the target position satisfies at least one or more of the following conditions: not collapsed, not on fire and not blocked by obstacles; according to the retraction encoder incremental information, a retraction operation is performed to reach the target position, and the next target retraction data is obtained according to the stored multiple motion state update data until the initial safety point is reached.

[0087] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0088] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A robot autonomous withdrawal method, characterized in that: The method is applied to a robot, and the method comprises: In response to receiving a start instruction of a target inspection task, obtaining location information of an initial safety point; Based on the location information of the initial safety point, start executing the target inspection task; In the process of executing the target inspection task, in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold, the motion state update data at the moment is acquired and stored; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; In response to satisfying the autonomous retraction condition, the target retraction data at the current moment is obtained according to the motion state update data stored at multiple moments; wherein the target retraction data includes the target position and the incremental information of the retraction encoder, and the target position at least satisfies one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; According to the incremental information of the retraction encoder, a retraction operation is performed to reach the target position, and the next target retraction data is continuously acquired according to the stored multiple motion state update data until the initial safety point is reached.

2. The autonomous withdrawal method of the robot according to claim 1, characterized in that: The obtaining of the location information of the initial safety point includes: Use laser radar to obtain point cloud information of the current location, and use camera equipment to obtain visual information of the current location; The position information of the initial safety point is obtained based on the point cloud information and the visual information by using a preset fusion algorithm.

3. The autonomous withdrawal method of the robot according to claim 1, characterized in that: The acquiring and storing the motion state update data at the moment includes: Obtaining encoder increment information at the moment according to the instantaneous value of the encoder at the moment and the instantaneous value of the encoder at the previous moment; Using an information fusion algorithm, according to the value of the inertial measurement unit IMU at the moment and the encoder increment information at the moment, the initial value of the position and posture of the robot at the moment is obtained; Using the iterative closest point ICP algorithm, the initial values ​​of the position and posture at the moment are corrected according to the point cloud information of the position at the moment obtained by using the laser radar and the visual information of the position at the moment obtained by using the camera equipment, and the corrected position, posture and encoder incremental information are stored as the motion state update data at the moment.

4. The autonomous withdrawal method of the robot according to claim 1, characterized in that: The response to satisfying the autonomous withdrawal condition includes: In the case where the communication signal of the robot is interrupted, confirming that the autonomous withdrawal condition is met; Alternatively, when the ambient temperature acquired by the ambient temperature detection device is higher than a preset temperature threshold, it is confirmed that the autonomous withdrawal condition is met.

5. The autonomous withdrawal method of the robot according to claim 1, characterized in that: The step of updating the motion state data stored at multiple moments to obtain the target withdrawal data at the current moment includes: Selecting, from the motion state update data at the multiple moments, the motion state update data closest to the current moment as candidate motion state update data; Determine whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by the laser radar and the visual information obtained by the camera device; wherein the retractable point is a position point that satisfies at least one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; In the case where the position included in the candidate motion state update data is a retractable point, using a preset inverse motion algorithm, according to the candidate motion state update data, obtain candidate retract data corresponding to the candidate motion state update data; wherein the candidate retract data includes a first candidate retract position and a candidate retract encoder increment; Obtaining the current position of the robot based on the point cloud information obtained by the laser radar and the visual information obtained by the camera device; Using the ICP algorithm, the current position of the robot is iteratively adjusted according to the candidate retraction encoder increment until the position adjustment termination condition is reached to obtain a second candidate retraction position; wherein the position adjustment termination condition includes: the difference between the first candidate retraction position and the second candidate retraction position is less than a preset error threshold, or a preset number of adjustments is reached; In response to the adjustment result that the difference between the first candidate retracement position and the second candidate retracement position is greater than the preset error threshold, the next motion state update data closest to the current moment is selected from the motion state update data stored at the plurality of moments as the candidate motion state update data, and the step of determining whether the position included in the candidate motion state update data is a retractable point based on the point cloud information obtained by using the laser radar and the visual information obtained by using the camera device and subsequent steps are repeatedly performed until the distance between the first candidate retracement position and the second candidate retracement position obtained by recalculation is less than the preset error threshold; In response to the adjustment result that the difference between the first candidate retracement position and the second candidate retracement position is less than the preset error threshold, the target retracement data is obtained according to the candidate retracement data.

6. The autonomous withdrawal method of the robot according to claim 5, characterized in that: The method further comprises: In the case where the position included in the candidate motion state update data is a non-retraceable point, a detour position is determined using the point cloud information obtained according to the laser radar and the visual information obtained by the camera device; wherein the detour position is located on a path, the path is perpendicular to the path from the position included in the candidate motion state update data to the current position of the robot, the path passes through the position included in the candidate motion state update data, and the detour position at least satisfies a combination of one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; The position in the motion status update data is updated according to the detour position.

7. A robot autonomous retraction device, characterized in that: The device comprises: A first acquisition module, configured to acquire location information of an initial safety point in response to receiving a start instruction of a target inspection task; An execution module, used for starting to execute the target inspection task based on the location information of the initial safety point; A second acquisition module is used for acquiring and storing the motion state update data at any moment in response to the change of the motion state of the robot at any moment relative to the previous moment being greater than a preset threshold during the execution of the target inspection task; wherein the motion state includes the position or posture of the robot, and the motion state update data includes the encoder increment information of the robot at the moment relative to the previous moment and the position and posture at the moment; A third acquisition module is used for obtaining target withdrawal data at a current moment according to the motion state update data stored at multiple moments in response to the autonomous withdrawal condition being met; wherein the target withdrawal data includes a target position and withdrawal encoder increment information, and the target position at least satisfies one or more of the following conditions: not collapsed, not on fire, and not blocked by obstacles; The retraction module is used to perform a retraction operation according to the retraction encoder incremental information to reach the target position, and continue to obtain the next target retraction data according to the stored multiple motion state update data until reaching the initial safety point.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the autonomous withdrawal method of the robot according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the autonomous withdrawal method of the robot according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the autonomous withdrawal method of the robot according to any one of claims 1 to 6 is implemented.

Images