Visual robot road sign positioning effective detection method

Abstract

The application discloses a visual robot road mark positioning effective detection method, comprising the following steps: acquiring a first positioning pose corresponding to a current road mark observed; acquiring a second positioning pose of a visual robot based on a visual sensor and an inertial sensor; judging whether the first positioning pose and the second positioning pose are the same, if yes, confirming that the first positioning pose corresponding to the current road mark is positioning effective, if not, calculating a first pose difference value of the first positioning pose and the second positioning pose and judging whether the first pose difference value is less than a preset difference threshold, if yes, confirming that the first positioning pose corresponding to the current road mark is positioning effective, if not, calculating a road mark positioning consistency degree of the current road mark and judging whether a road mark positioning consistency degree requirement is met, if yes, the first positioning pose corresponding to the current road mark is positioning effective, if not, the first positioning pose corresponding to the current road mark is positioning ineffective. The method improves the positioning reliability and accuracy of the visual robot based on the road mark.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of robot positioning, in particular to a visual robot landmark positioning effective detection method. BACKGROUND

[0002] The current positioning and navigation technology of mobile robots mainly realizes through vision, laser or inertial navigation technology, etc., wherein the positioning method based on vision is widely applied in indoor positioning of mobile robots. The method for realizing positioning and navigation of mobile robots based on vision includes positioning based on landmark identification, and the positioning method based on landmark identification in the prior art mainly realizes by identifying landmarks, directly taking the positioning information corresponding to the landmarks as the positioning information of the visual robot. This method has the problem that the positioning error of the landmarks directly leads to the positioning error of the visual robot, and the positioning reliability is low. SUMMARY

[0003] To solve the above problems, the present application provides a visual robot landmark positioning effective detection method, which combines the visual robot positioning pose obtained by the visual-inertial system (VINS) to detect whether the landmark positioning is effective, so as to determine whether to use landmark positioning, avoid the situation that the positioning error of the landmarks leads to the positioning error of the visual robot, and ensure the positioning reliability and accuracy of the visual robot based on landmarks. The specific technical scheme of the present application is as follows:

[0004] A visual robot landmark positioning effective detection method, the visual robot landmark positioning effective detection method comprises: controlling the visual robot to obtain a first positioning pose corresponding to a current landmark observed; controlling the visual robot to obtain a second positioning pose of the visual robot based on a visual sensor and an inertial sensor; judging whether the first positioning pose and the second positioning pose are the same; if the first positioning pose and the second positioning pose are the same, confirming that the first positioning pose corresponding to the current landmark is positioned effectively; if the first positioning pose and the second positioning pose are not the same, calculating a first pose difference value of the first positioning pose and the second positioning pose; judging whether the first pose difference value of the first positioning pose and the second positioning pose is less than a preset difference threshold value; if yes, confirming that the first positioning pose corresponding to the current landmark is positioned effectively; if no, calculating a landmark positioning consistency degree of the current landmark according to all recorded observed landmarks, judging whether the landmark positioning consistency degree of the current landmark meets a landmark positioning consistency degree requirement, if the landmark positioning consistency degree of the current landmark meets the landmark positioning consistency degree requirement, confirming that the first positioning pose corresponding to the current landmark is positioned effectively, if the landmark positioning consistency degree of the current landmark does not meet the landmark positioning consistency degree requirement, confirming that the first positioning pose corresponding to the current landmark is not positioned effectively.

[0005] Further, the method for calculating the current landmark positioning consistency degree according to all recorded observed landmarks specifically comprises: performing a positioning consistency matching of the current landmark with all recorded observed landmarks, counting the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark, and confirming the current landmark positioning consistency degree based on the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark.

[0006] Further, the method for performing the positioning consistency matching of the current landmark with all recorded observed landmarks specifically comprises: step 1: selecting one observed landmark which has not been subjected to the positioning consistency matching from all recorded observed landmarks as a current positioning consistency matching landmark; step 2: acquiring the observation time of the current positioning consistency matching landmark and acquiring the positioning pose corresponding to the current positioning consistency matching landmark; step 3: calculating a second pose difference value between the first positioning pose corresponding to the current landmark and the positioning pose corresponding to the current positioning consistency matching landmark; step 4: calculating a first displacement distance of the visual robot in a time period from the observation time of the current positioning consistency matching landmark to the observation time of the current landmark according to the inertial sensor of the visual robot; step 5: judging whether the second pose difference value and the first displacement distance are the same, if the second pose difference value and the first displacement distance are the same, it is confirmed that the current positioning consistency matching landmark is consistent with the current landmark in positioning, if the second pose difference value and the first displacement distance are not the same, it is confirmed that the current positioning consistency matching landmark is inconsistent with the current landmark in positioning; repeatedly performing the above steps 1 to 5 until there is no observed landmark which has not been subjected to the positioning consistency matching in all recorded observed landmarks.

[0007] Further, the method for counting the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark specifically comprises: sequentially traversing and judging whether each observed landmark in all recorded observed landmarks is consistent with the current landmark in positioning, if the currently traversed and judged observed landmark is consistent with the current landmark in positioning, the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark is increased by 1, if the currently traversed and judged observed landmark is inconsistent with the current landmark in positioning, the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark remains unchanged, and the above steps are repeatedly performed, and when all recorded observed landmarks are sequentially traversed and judged, the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark is acquired.

[0008] Further, the number of landmarks existing in all recorded observed landmarks which are consistent with the positioning of the current landmark and the current landmark positioning consistency degree have a positive correlation.

[0009] Further, the landmark consistency requirement refers to requiring the number of landmarks existing in the recorded observed landmarks and consistent with the current landmark positioning to reach a first quantity threshold.

[0010] Further, the first quantity threshold is positively correlated with the first pose difference of the first positioning pose and the second positioning pose.

[0011] Further, when all the recorded observed landmarks are sequentially and completely judged, and the number of landmarks existing in the recorded observed landmarks and consistent with the current landmark positioning is zero, the current landmark is recorded as an observed landmark, and the recorded observed landmarks are updated.

[0012] Further, the method for recording the current landmark as an observed landmark specifically comprises: recording the first positioning pose corresponding to the current landmark, and recording the observation time of the current landmark.

[0013] Further, when the observed landmark is consistent with the current landmark positioning, the observation time of the observed landmark is updated as the observation time of the current landmark.

[0014] Further, when the time length of the observation time of the observed landmark from the current time reaches a first time threshold, the observed landmark is deleted from the recorded observed landmarks, and the recorded observed landmarks are updated.

[0015] The visual robot positioning method provided by the application considers the landmark consistency of the positioning pose of the current landmark when the positioning pose of the current landmark observed by the visual robot is different from the positioning pose of the visual robot acquired based on the visual inertial system, determines that the positioning of the landmark that does not meet the landmark consistency requirement is invalid, avoids the problem of visual robot positioning failure caused by incorrect identification of the landmark, and makes the positioning accuracy of the visual robot based on the landmark higher and more reliable. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The flowchart of the visual robot landmark positioning effective detection method of the first embodiment of the application is shown.

[0017] Figure 2 The flowchart of the positioning consistency matching method of the third embodiment of the application is shown. DETAILED DESCRIPTION

[0018] In order to make the objects, technical solutions and advantages of the present application clearer, the following will describe and explain the present application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described below are only used to explain the present application and should not be used to limit the present application. In addition, it can also be understood that for those skilled in the art, some design, manufacture or production changes on the technical content disclosed in the present application are only conventional technical means and should not be understood as the insufficiency of the present application.

[0019] Unless otherwise defined, the technical terms or scientific terms involved in the present application should be the usual meaning understood by those skilled in the art to which the present application belongs. The terms "comprise", "include", "have" and any variations thereof involved in the present application are intended to cover non-exclusive inclusion, such as: a process, method, system product or device comprising a series of steps or modules, which is not limited to the listed steps or modules, but can also include steps or modules not listed, or can also include other steps or modules inherent to the process, method, product or device.

[0020] As a preferred embodiment of the present application, the first embodiment of the present application provides a visual robot road sign positioning effective detection method, and the road sign establishment method can be but is not limited to being established according to environmental landmarks, or being artificially configured, or being established based on existing road sign distribution information; the visual robot refers to a mobile robot for positioning and navigation based on a visual sensor, and the visual sensor can be but is not limited to a monocular camera, a binocular camera, a depth camera or a fisheye camera, etc. sensor with visual image acquisition function.

[0021] As shown in Figure 1 , the visual robot road sign positioning effective detection method specifically comprises:

[0022] controlling the visual robot to obtain a first positioning pose corresponding to a current road sign observed at present; it should be noted that there is a one-to-one correspondence between each road sign and a positioning pose, and the positioning pose corresponding to each road sign is recorded or configured when it is established.

[0023] controlling the visual robot to obtain a second positioning pose of the visual robot at present based on the visual sensor and the inertial sensor; specifically, the second positioning pose of the visual robot at present is calculated and obtained based on a visual inertial system composed of the visual sensor and the inertial sensor mounted on the body of the visual robot.

[0024] judging whether the first positioning pose corresponding to the current road sign and the second positioning pose of the visual robot at present are the same;

[0025] If the first positioning pose corresponding to the current landmark is the same as the second positioning pose of the visual robot, it is determined that the first positioning pose corresponding to the current landmark is valid; it should be noted that if the first positioning pose corresponding to the current landmark is valid, it means that the current landmark can be used as a positioning reference of the visual robot.

[0026] If the first positioning pose corresponding to the current landmark is different from the second positioning pose of the visual robot, a first pose difference between the first positioning pose and the second positioning pose is calculated, and it is determined whether the first pose difference is less than a preset difference threshold; if the first pose difference is less than the preset difference threshold, it is determined that the first positioning pose corresponding to the current landmark is valid; otherwise, if the first difference is greater than or equal to the preset difference threshold, a landmark positioning consistency of the current landmark is calculated according to all recorded observed landmarks, and it is determined whether the landmark positioning consistency of the current landmark meets a landmark positioning consistency requirement; if the landmark positioning consistency of the current landmark meets the landmark positioning consistency requirement, it is determined that the first positioning pose corresponding to the current landmark is valid; otherwise, if the landmark positioning consistency of the current landmark does not meet the landmark positioning consistency requirement, it is determined that the first positioning pose corresponding to the current landmark is invalid. Specifically, the all recorded observed landmarks are landmarks recorded when previously observed landmarks meet a recording condition, which are used as a reference for the landmark positioning consistency of the current landmark.

[0027] The visual robot landmark positioning validity detection method provided in the embodiment considers the landmark consistency of the positioning pose of the current landmark when the positioning pose of the current landmark observed is different from the positioning pose of the visual robot obtained based on the visual-inertial system, and determines that the positioning of the landmark that does not meet the landmark consistency requirement is invalid, thereby avoiding the problem of positioning failure of the visual robot caused by incorrect identification of the landmark, and making the positioning accuracy of the visual robot based on the landmark higher and more reliable.

[0028] Based on the first embodiment, as a preferred embodiment of the present application, the method for calculating the positioning consistency degree of the current landmark according to all the recorded observed landmarks in the second embodiment of the present application specifically comprises: performing a positioning consistency matching on the current landmark and all the recorded observed landmarks, counting the number of landmarks existing in all the recorded observed landmarks which are consistent with the positioning of the current landmark, and obtaining the positioning consistency degree of the current landmark based on the number of landmarks existing in all the recorded observed landmarks which are consistent with the positioning of the current landmark. Specifically, it needs to be noted that the positioning consistency degree of the landmark is positively correlated with the number of landmarks existing in all the recorded observed landmarks which are consistent with the positioning of the current landmark, and the more the number of landmarks existing in all the recorded observed landmarks which are consistent with the positioning of the current landmark, the higher the positioning consistency degree of the current landmark. Therefore, in the method provided in the embodiment, the number of landmarks existing in all the recorded observed landmarks which are consistent with the positioning of the current landmark is calculated, so as to confirm the positioning consistency degree of the current landmark. The higher the positioning consistency degree of the current landmark, the higher the reliability of the first positioning pose corresponding to the current landmark. The present application understands the reliability of the current landmark by calculating the positioning consistency degree of the current landmark, and realizes the grasping of the landmark situation.

[0029] Based on the above embodiments, as a preferred embodiment of the present application, as shown in Figure 2 The method for performing the positioning consistency matching on the current landmark and all the recorded observed landmarks in the third embodiment of the present application specifically comprises:

[0030] Step 1: selecting one observed landmark which has not been subjected to the positioning consistency matching from all the recorded observed landmarks as a current positioning consistency matching landmark. Specifically, an observed landmark which has not been subjected to the positioning consistency matching is selected from all the recorded observed landmarks as a current positioning consistency matching landmark, that is, the observed landmark is selected for the positioning consistency matching. By judging the observed landmarks existing in all the recorded observed landmarks which have not been subjected to the positioning consistency matching, the traversal of the positioning consistency matching on all the recorded observed landmarks is realized, and no observed landmark is missed, so as to ensure the accuracy of the landmark consistency matching. It needs to be noted that the all the recorded observed landmarks refer to the landmarks observed by the visual robot in the past, and each time a landmark meeting the recording condition is encountered, the landmark is recorded into all the recorded observed landmarks. Specifically, the recording condition can be but is not limited to the condition that the landmark has never been observed, and the like, which can limit the non-repetition of the recorded observed landmarks.

[0031] Step 2: obtaining the observation time of the current positioning consistent matching road sign and obtaining the positioning pose corresponding to the current positioning consistent matching road sign; specifically, there is a one-to-one correspondence between each road sign and an observation time and a positioning pose, the observation time refers to the time when the road sign is observed, and it should be noted that the observation time corresponding to each road sign is variable and not constant, and when a road sign is observed again, the observation time corresponding to the road sign is updated to the current time; the positioning pose refers to the positioning pose recorded at the time when the road sign is observed for the first time, and the positioning pose corresponding to each road sign does not change with the number of observations.

[0032] Step 3: calculating a second pose difference value between the first positioning pose corresponding to the current road sign and the positioning pose corresponding to the current positioning consistent matching road sign;

[0033] Step 4: calculating a first displacement distance of the vision robot in a time period from the observation time of the current positioning consistent matching road sign to the observation time of the current road sign according to the inertial sensor of the vision robot; specifically, the inertial sensor calculates the displacement distance of the vision robot in real time based on the code disc and the gyroscope during the movement of the vision robot.

[0034] Step 5: judging whether the second pose difference value and the first displacement distance are the same, if the second pose difference value and the first displacement distance are the same, it is confirmed that the current positioning consistent matching road sign and the current road sign are positioned consistently, otherwise, if the second pose difference value and the first displacement distance are different, it is confirmed that the current positioning consistent matching road sign and the current road sign are not positioned consistently; specifically, whether the second pose difference value and the first displacement distance are the same is judged to determine whether the current positioning consistent matching road sign and the current road sign are positioned consistently.

[0035] Repeat steps 1 to 5 above until there is no observed road sign that has not been positioned consistently among all recorded observed road signs.

[0036] The embodiment determines whether there is a road sign that is positioned consistently with the current road sign among all recorded observed road signs by sequentially traversing the positioning consistent matching results of the observed road sign and the current road sign, so as to realize the detection of the positioning reliability of the current road sign.

[0037] Based on the above embodiment, as a more preferred embodiment of the present application, the method for counting the number of road signs that are positioned consistently with the current road sign among all recorded observed road signs in the fourth embodiment of the present application specifically comprises:

[0038] sequentially traversing each of the observed landmarks in the recorded observed landmarks to determine whether each of the observed landmarks is consistent with the current landmark in positioning; specifically, the method of determining whether each of the observed landmarks in the recorded observed landmarks is consistent with the current landmark in positioning can be, but is not limited to, the method described in the third embodiment of the present application.

[0039] If the observed landmark currently determined in the traversal is consistent with the current landmark in positioning, the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks is increased by 1;

[0040] If the observed landmark currently determined in the traversal is not consistent with the current landmark in positioning, the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks remains unchanged;

[0041] The above steps are repeated until the recorded observed landmarks are all traversed sequentially, and the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks is obtained.

[0042] The embodiment sequentially traverses the recorded observed landmarks and the current landmark to determine the consistency in positioning, and increases the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks by 1 each time the observed landmark is consistent with the current landmark in positioning, and otherwise, keeps the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks unchanged, thereby accurately counting the number of the observed landmarks consistent with the current landmark in positioning in the recorded observed landmarks.

[0043] Based on the above embodiments, as a more preferred embodiment of the present application, in the fifth embodiment of the present application, the method of matching the current landmark with the recorded observed landmarks in positioning to count the number of the observed landmarks consistent with the current landmark in positioning, specifically includes:

[0044] Step A: selecting one of the recorded observed landmarks that has not been matched in positioning as a current landmark for matching in positioning; specifically, selecting one of the recorded observed landmarks that has not been matched in positioning as a current landmark for matching in positioning, i.e., selecting the observed landmark for matching in positioning, thereby achieving the traversal of matching in positioning for all the recorded observed landmarks without missing any observed landmark, and ensuring the accuracy of the matching in positioning.

[0045] Step B: obtaining the observation time of the current landmark for matching in positioning and obtaining the corresponding positioning pose of the current landmark for matching in positioning;

[0046] Step C: calculating a second pose difference value of the first positioning pose corresponding to the current landmark and the positioning pose corresponding to the current positioning consistent matching landmark;

[0047] Step D: calculating a first displacement distance of the visual robot in a time period from the observation time of the current positioning consistent matching landmark to the observation time of the current landmark according to the inertial sensor of the visual robot;

[0048] Step E: judging whether the second pose difference value and the first displacement distance are the same, if the second pose difference value and the first displacement distance are the same, it is confirmed that the current positioning consistent matching landmark and the current landmark are consistent in positioning, the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks is increased by 1, otherwise, if the second pose difference value and the first displacement distance are different, it is confirmed that the current positioning consistent matching landmark and the current landmark are inconsistent in positioning, the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks is kept unchanged;

[0049] The above steps A to E are repeated until there is no observed landmark not subjected to positioning consistent matching existing in the recorded observed landmarks.

[0050] The embodiment realizes the acquisition of the positioning consistency degree of the current landmark by performing positioning consistent matching between the recorded observed landmarks and the current landmark one by one respectively and counting the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks according to the positioning consistent matching result.

[0051] It should be noted that the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks and the positioning consistency degree of the current landmark are in a positive correlation. When the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks is more, it means that the positioning consistency degree of the current landmark is higher, otherwise, when the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks is less, it means that the positioning consistency degree of the current landmark is lower. The positioning consistency degree of the current landmark is higher, which means that the positioning accuracy of the current landmark is higher and the first positioning pose corresponding to the current landmark is higher in reliability, and the positioning consistency degree of the current landmark is lower, which means that the first positioning pose corresponding to the current landmark is lower in reliability and the accuracy of the current landmark needs to be further tested.

[0052] Based on the above embodiment, as a more preferred embodiment of the present application, the landmark consistency degree requirement in the sixth embodiment of the present application is configured to require the number of the landmarks consistent in positioning with the current landmark existing in the recorded observed landmarks to reach a first number threshold. Specifically, the first number threshold is a variable adjustable value for limiting whether the consistency degree of the current landmark reaches the requirement.

[0053] Preferably, the first quantity threshold value is in positive correlation with the first pose difference value of the first positioning pose and the second positioning pose. When the first pose difference value of the first positioning pose and the second positioning pose is large, the first quantity threshold value is adjusted to be large accordingly, and vice versa. For example, when the first pose difference value of the first positioning pose and the second positioning pose is small, the first quantity threshold value is set to 2, and only when the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning is 2 or more, the landmark consistency degree of the current landmark meets the landmark consistency degree requirement. When the first pose difference value of the first positioning pose and the second positioning pose is large, the first quantity threshold value is set to 5, and only when the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning is 5 or more, the landmark consistency degree of the current landmark meets the landmark consistency degree requirement. If the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning is less than 5, the landmark consistency degree of the current landmark does not meet the landmark consistency degree requirement.

[0054] It should be noted that the first quantity threshold value is configured to be different values according to the size of the first pose difference value. The first quantity threshold value can be, but is not limited to, an integer greater than 0.

[0055] In the embodiment, the landmark consistency degree of the current landmark is linked to the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning, so that the landmark consistency degree of the current landmark can be directly grasped through the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning. Meanwhile, the landmark consistency degree requirement is in positive correlation with the first pose difference value, so that whether the landmark consistency degree requirement needs to be adjusted according to the size of the preliminary pose difference value during positioning can be determined, and the landmark positioning reliability is improved through multiple flexible determination methods.

[0056] Based on the above embodiment, as a more preferred embodiment of the present application, in the seventh embodiment of the present application, when all recorded observed landmarks are sequentially and completely judged, and the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning is zero, it indicates that the current landmark has never been observed, and the current landmark is recorded as an observed landmark, and all recorded observed landmarks are updated. In the embodiment, the current landmark with the number of the observed landmarks existing in all recorded observed landmarks and consistent with the current landmark positioning being zero after all recorded observed landmarks are completely judged is recorded as an observed landmark, so that the unobserved landmark is recorded in the observed landmark, and all recorded observed landmarks are increased, and the comprehensive reliability of all recorded observed landmark data is ensured.

[0057] Based on the above-mentioned seventh embodiment, as a preferred embodiment of the present application, in the eighth embodiment of the present application, the method of recording the current landmark as an observed landmark is specifically recording the first positioning pose corresponding to the current landmark and recording the observation time of the current landmark. It should be noted that the information of the recorded observed landmarks used in the present application is mainly the positioning pose corresponding to the observed landmark and the observation time corresponding to the observed landmark. In actual application process, the method of recording the current landmark as an observed landmark can also include recording other current landmark related information. It should be noted that the observation time of the current landmark is changeable and updateable.

[0058] Based on the above-mentioned embodiment, as a preferred embodiment of the present application, in the ninth embodiment of the present application, in the process of sequentially traversing and judging whether each observed landmark of the recorded observed landmarks is consistent with the current landmark in positioning, if the current traversed and judged observed landmark is consistent with the current landmark in positioning, the observation time corresponding to the current traversed and judged observed landmark is updated to the observation time of the current landmark. Since the current landmark is consistent with the observed landmark in positioning, it means that the visual robot observes the observed landmark again at the current time. Therefore, the observation time of the current landmark is updated to the observation time corresponding to the observed landmark, which indicates that the observed landmark is observed again. In this embodiment, the observation time corresponding to the observed landmark is updated according to the consistency judgment result of the observed landmark and the current landmark in positioning, so that the observation time corresponding to the recorded observed landmarks is kept as the latest observation time.

[0059] Based on the above-mentioned embodiment, as a preferred embodiment of the present application, in the tenth embodiment of the present application, when the time length from the observation time corresponding to the observed landmark to the current time reaches the first time threshold, the observed landmark is deleted from the recorded observed landmarks, and the recorded observed landmarks are updated. Specifically, the first time threshold is used to limit the maximum time length that the observed landmark needs to be observed again. The first time threshold is set according to the actual application scene, moving speed and other factors of the visual robot. When the time length that the observed landmark is not observed again exceeds the maximum time length, it is considered that the observed landmark is invalid, which needs to be deleted from the recorded observed landmarks, so as to reduce the calculation burden of the visual robot algorithm and improve the positioning efficiency of the visual robot.

[0060] Obviously, the above embodiments are only a part of the embodiments of the present application, and not all the embodiments, and the technical solutions of each embodiment can be combined with each other. In the above embodiments of the present application, the description of each embodiment is focused on, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments. In the several embodiments provided by the present application, it should be understood that the disclosed technology can be implemented in other ways. In addition, it should be noted that, in the above specific embodiments, each specific technical feature described can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present application will not be described again.

[0061] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A visual robot road sign positioning effective detection method, characterized in that, The visual robot road marker positioning effective detection method comprises: controlling the visual robot to acquire a first positioning pose corresponding to a current road marker observed; controlling the visual robot to acquire a second positioning pose of the visual robot based on a visual sensor and an inertial sensor; judging whether the first positioning pose and the second positioning pose are the same; if the first positioning pose and the second positioning pose are the same, confirming that the first positioning pose corresponding to the current road marker is positioned effectively; if the first positioning pose and the second positioning pose are not the same, calculating a first pose difference value of the first positioning pose and the second positioning pose; judging whether the first pose difference value of the first positioning pose and the second positioning pose is less than a preset difference threshold value; if yes, confirming that the first positioning pose corresponding to the current road marker is positioned effectively; if no, calculating a road marker positioning consistency degree of the current road marker according to all recorded observed road markers, judging whether the road marker positioning consistency degree of the current road marker meets a road marker positioning consistency degree requirement, if the road marker positioning consistency degree of the current road marker meets the road marker positioning consistency degree requirement, confirming that the first positioning pose corresponding to the current road marker is positioned effectively, and if the road marker positioning consistency degree of the current road marker does not meet the road marker positioning consistency degree requirement, confirming that the first positioning pose corresponding to the current road marker is not positioned effectively; wherein the method for calculating the road marker positioning consistency degree of the current road marker according to all recorded observed road markers specifically comprises: performing a certain positioning consistent matching between the current road marker and all recorded observed road markers, counting the number of road markers that are positioned consistently with the current road marker among all recorded observed road markers, and confirming the road marker positioning consistency degree of the current road marker based on the number of road markers that are positioned consistently with the current road marker among all recorded observed road markers; wherein the method for performing the certain positioning consistent matching between the current road marker and all recorded observed road markers specifically comprises: Step 1: selecting one observed road marker that has not been subjected to positioning consistent matching from all recorded observed road markers as a current positioning consistent matching road marker; Step 2: acquiring an observation time of the current positioning consistent matching road marker and acquiring a positioning pose corresponding to the current positioning consistent matching road marker; Step 3: calculating a second pose difference value of the first positioning pose corresponding to the current road marker and the positioning pose corresponding to the current positioning consistent matching road marker; Step 4: calculating a first displacement distance of the visual robot in a time period from the observation time of the current positioning consistent matching road marker to the observation time of the current road marker according to the inertial sensor of the visual robot; Step 5: judging whether the second pose difference value and the first displacement distance are the same, if the second pose difference value and the first displacement distance are the same, confirming that the current positioning consistent matching road marker and the current road marker are positioned consistently, and if the second pose difference value and the first displacement distance are not the same, confirming that the current positioning consistent matching road marker and the current road marker are not positioned consistently; repeating the above steps 1 to 5 until there is no observed road marker that has not been subjected to positioning consistent matching among all recorded observed road markers.

2. The visual robot road marker positioning efficient detection method according to claim 1, characterized in that, The method for counting the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position specifically comprises: sequentially determining whether each observed landmark in the recorded observed landmarks is consistent with the current landmark position, if the currently determined observed landmark is consistent with the current landmark position, the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position is increased by 1, if the currently determined observed landmark is not consistent with the current landmark position, the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position remains unchanged, and the above steps are repeated, and when all the recorded observed landmarks are sequentially determined, the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position is obtained.

3. The visual robot road marker positioning efficient detection method according to claim 2, characterized in that, The number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position is positively correlated with the consistency degree of the current landmark position.

4. The visual robot road marker positioning efficient detection method according to claim 3, characterized in that, The landmark consistency degree requirement is that the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position reaches a first number threshold.

5. The visual robot road marker positioning efficient detection method according to claim 4, characterized in that, The first number threshold is positively correlated with the first pose difference between the first positioning pose and the second positioning pose. 6.The visual robot road marker positioning effective detection method according to claim 2, characterized in that, When all the recorded observed landmarks are sequentially determined, and the number of observed landmarks in the recorded observed landmarks that are consistent with the current landmark position is zero, the current landmark is recorded as an observed landmark, and the recorded observed landmarks are updated.

7. The visual robot road marker positioning efficient detection method according to claim 6, characterized in that, The method for recording the current landmark as an observed landmark specifically comprises: recording the first positioning pose corresponding to the current landmark, and recording the observation time of the current landmark.

8. The visual robot road marker positioning effective detection method according to claim 7, characterized in that, When the observed landmark is consistent with the current landmark position, the observation time of the observed landmark is updated to the observation time of the current landmark.

9. The visual robot road marker positioning efficient detection method according to claim 8, characterized in that, When the time length from the observation time of the observed landmark to the current time reaches a first time threshold, the observed landmark is deleted from the recorded observed landmarks, and the recorded observed landmarks are updated.

Images