Mechanical arm motion trail optimization method and system, storage medium and electronic equipment

Abstract

The invention provides a mechanical arm motion trail optimization method and system, a storage medium and electronic equipment, and relates to the technical field of mechanical arm control, and the method comprises the steps that a first latest key area for interaction between a mechanical arm and a user is recognized; a first local track before the mechanical arm enters the first latest key area and a second local track after the mechanical arm enters the first latest key area are determined from the first current future motion track of the mechanical arm; and based on the first interaction condition of interaction between the mechanical arm and the user when the mechanical arm moves along the first local track, optimization replacement processing is conducted on the second local track. Therefore, the mechanical arm can adapt to the emergency when moving along the second local track subjected to optimization replacement processing in the future, flexible and intelligent response can be made when the user demand changes suddenly, the normal activity of the user is prevented from being interfered, the interaction process is ensured to be smooth, and the user experience is improved to a great extent.

Description

Technical Field

[0001] The present invention relates to the field of robot arm control technology, and in particular to a robot arm motion trajectory optimization method, system, storage medium and electronic equipment. Background Art

[0002] With the rapid development of robotic arm technology, its application scenarios have covered many fields such as industrial production, home services, medical health, etc. In these applications, the robotic arm needs to interact with the user in real time, such as moving objects, delivering objects, and fixing objects.

[0003] However, the motion trajectory planning of the robot arm often needs to be dynamically adjusted according to the actual environment and user needs. Existing technologies generally rely on preset motion paths and fixed trajectory planning, and lack a flexible response mechanism to changes in real-time user needs. Although in some scenarios, the robot arm already has a certain degree of automatic trajectory planning capabilities, most of the current robot arms still find it difficult to respond flexibly and intelligently when interacting with users, especially when user needs suddenly change (such as when the user suddenly turns on the TV or engages in other activities). This may cause the movement of the robot arm to interfere with the user's normal activities (such as blocking the TV screen, affecting the line of sight, or preventing the user from performing other operations), resulting in an unsmooth interaction process, which in turn greatly affects the user experience.

[0004] Therefore, there is an urgent need for a more flexible trajectory optimization method that can optimize the motion trajectory of the robot arm in an intelligent way so that it can automatically adapt to sudden changes in user needs and avoid unnecessary conflicts during the interaction process. Summary of the invention

[0005] One of the purposes of the present invention is to provide a method for optimizing the motion trajectory of a robotic arm. When an emergency occurs during the interaction between the robotic arm and a user, a first latest key area that requires local trajectory optimization is identified, and based on a first interaction situation of the robotic arm interacting with the user when the robotic arm moves along a first local trajectory before entering the first latest key area, a second local trajectory entering the first latest key area is optimized and replaced, so that the robotic arm can adapt to the emergency when it moves along the second local trajectory after the optimization and replacement in the future, so as to achieve flexible and intelligent response when user needs suddenly change, avoid interfering with the user's normal activities, ensure a smooth interaction process, and greatly improve the user experience.

[0006] An embodiment of the present invention provides a method for optimizing a robot arm motion trajectory, comprising: Identifying a first and latest critical area for interaction between the robotic arm and the user; Determine respectively from a first current future motion trajectory of the robot arm a first local trajectory before entering a first latest critical area and a second local trajectory entering the first latest critical area; Based on a first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory, an optimization replacement process is performed on the second local trajectory.

[0007] Optionally, the identifying the first latest key area for interaction between the robotic arm and the user includes: Obtain the user's field of view change sequence after the robot arm last interacted with the user; Based on the appearance of the robot arm in each first field of view in the field of view change sequence, an appearance curve is drawn; When more than N target peaks higher than the first peak appear successively after the first peak in the appearance degree curve, the first and last peaks of the target peak are intercepted from the field of view change sequence, and each of them corresponds to a local field of view sequence between the first fields of view; wherein N is a positive integer greater than or equal to 2; Acquire a set of target objects that appear in every second field of view in the local field of view sequence; Matching the target object set with multiple standard object sets respectively; When the match is met, the key object indication rule of the matching standard object set is obtained; Based on the key object indication rule, determine the key objects and corresponding optimization categories from the target object set; When the optimization category is active optimization, the area of ​​the first visual field corresponding to the last occurrence of the key object in the first visual field in the trough of the appearance curve is taken as the first latest key area; When the optimization category is passive optimization, the area of ​​the overlapping fields of view between the target peaks corresponding to the first fields of view is taken as the first latest key area.

[0008] Optionally, the optimizing and replacing processing of the second partial trajectory based on a first interaction situation of the robot arm interacting with the user when the robot arm moves along the first partial trajectory includes: Performing feature description processing on the first interaction situation to obtain a first feature description vector; When the optimization category is active optimization, determining the active behavior corresponding to the first feature description vector from the active behavior library; Performing an optimization replacement process on the second local trajectory, so that the robot arm can exhibit active behavior in the first latest key area when moving along the second local trajectory after the optimization replacement process; When the optimization category is passive optimization, determining the passive behavior corresponding to the first feature description vector from the passive behavior library; The second local trajectory is optimized and replaced, so that the robot arm can show passive behavior in any area outside the first latest key area when moving along the second local trajectory after the optimization and replacement processing.

[0009] Optionally, the step of drawing an appearance curve based on the appearance of the robot arm in each first field of view in the field of view change sequence comprises: Based on the curve drawing template, an appearance curve is drawn according to the appearance of the robot arm in each first field of view in the field of view change sequence; The horizontal axis of the template curve coordinate system in the curve drawing template is the order of field of view change, and the vertical axis of the template curve coordinate system is the magnitude of the appearance.

[0010] Optionally, after performing the optimization and replacement processing on the second partial trajectory based on the first interaction situation between the robot arm and the user when the robot arm moves along the first partial trajectory, the method further includes: After the robot arm finishes moving along the second local trajectory after the optimization replacement process, if other users stay in the first latest key area for more than a first preset time, identifying the second latest key area for interaction between the robot arm and the user and other users; Determine respectively from the second current future motion trajectory of the robot arm a third local trajectory before entering the second latest key area and a fourth local trajectory entering the second latest key area; Based on a second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory, the fourth partial trajectory is optimized and replaced.

[0011] Optionally, the identifying the second latest key area for interaction between the robot arm and the user or other users includes: When the key object does not appear in the third visual field of the user and other users for more than a second preset time period, an overlapping area of ​​the third visual field of the user and other users that overlaps for more than a third preset time period is taken as the second latest key area.

[0012] Optionally, the optimizing and replacing processing on the fourth partial trajectory based on a second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory includes: Performing feature description processing on the second interaction situation to obtain a second feature description vector; Determining the interaction triggering degree corresponding to the second feature description vector from the interaction triggering degree library; When the interaction trigger degree exceeds the trigger degree threshold, the robot arm is planned to continuously dock with the key object; The fourth local trajectory is optimized and replaced so that the robot arm can exhibit docking behavior in the second latest key area when moving along the fourth local trajectory after the optimization and replacement processing.

[0013] An embodiment of the present invention provides a robot arm motion trajectory optimization system, comprising: A latest key area recognition module, used to recognize the first latest key area for interaction between the robot arm and the user; A local trajectory determination module, used to determine a first local trajectory before entering a first latest key area and a second local trajectory entering the first latest key area from a first current future motion trajectory of the robot arm; The trajectory optimization and replacement processing module is used to optimize and replace the second local trajectory based on a first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory.

[0014] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and a processor executes the computer program to implement any of the above methods.

[0015] An embodiment of the present invention provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any one of the methods described above.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0017] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A schematic diagram of a method for optimizing the motion trajectory of a robot arm according to an embodiment of the present invention; Figure 2 Schematic diagram of a robot arm motion trajectory optimization system in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0020] The embodiment of the present invention provides a method for optimizing the motion trajectory of a robot arm. Figure 1 As shown, including: S1, identifying the first latest key area for interaction between the robot arm and the user; S2, determining respectively from the first current future motion trajectory of the robot arm a first local trajectory before entering the first latest key area and a second local trajectory entering the first latest key area; S3. Based on a first interaction situation between the robot arm and the user when the robot arm moves along the first partial trajectory, optimize and replace the second partial trajectory.

[0021] The working principle and beneficial effects of the above technical solution are: The interaction between the robot arm and the user refers to the process in which the robot arm provides services to the user and interacts with the user, for example: the robot arm carries the items required by the user and delivers them to the user, the robot arm temporarily holds an item for the user, etc. When the robot arm is working, it will adaptively plan the future motion trajectory according to its own working conditions in real time. The first current future motion trajectory is the trajectory of the future movement of the robot arm after the moment of identifying the first latest key area.

[0022] When an emergency occurs during the interaction between the robot and the user, the first latest key area will be generated. The second local trajectory entering the first latest key area needs to be optimized and replaced according to the emergency. The emergency will further develop when the robot moves along the first local trajectory. Therefore, the first interaction situation can be used to optimize and replace the second local trajectory. Specifically, for example: when a user temporarily turns on the TV to watch a movie, in order to avoid blocking the user's line of sight, the first latest key area is the field of view of the user when watching TV. The first interaction situation is that there is no interaction between the robot and the user, which further indicates that the user wants to watch TV at ease at this time, so the second local trajectory is optimized and replaced. The goal of the optimization and replacement is to prevent the robot from moving in the first latest key area in the future.

[0023] When an emergency occurs during the interaction between the robot arm and the user, the present application identifies the first latest key area that requires local trajectory optimization, and based on the first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory before entering the first latest key area, optimizes and replaces the second local trajectory entering the first latest key area, so that the robot arm can adapt to the emergency when it moves along the second local trajectory after the optimization and replacement in the future, thereby achieving flexible and intelligent response when user needs suddenly change, avoiding interference with the user's normal activities, ensuring a smooth interaction process, and greatly improving the user experience.

[0024] In the application scenario of the above-mentioned embodiment of the present application, when an emergency occurs during the interaction between the robot arm and the user, the recognition accuracy of the first latest key area directly determines the effect of the optimization and replacement processing of the second local trajectory entering it. If the first latest key area is too large, the system will consume too many resources for trajectory optimization. If the first latest key area is too small, the trajectory optimization will not be able to enable the interaction between the robot arm and the user to flexibly adapt to the unexpected situations that arise.

[0025] Therefore, to solve the above problem, in one embodiment, the step of identifying the first latest key area for interaction between the robot arm and the user includes: Obtain the user's field of view change sequence after the robot arm last interacted with the user; Based on the appearance of the robot arm in each first field of view in the field of view change sequence, an appearance curve is drawn; When more than N target peaks higher than the first peak appear successively after the first peak in the appearance degree curve, the first and last peaks of the target peak are intercepted from the field of view change sequence, and each of them corresponds to a local field of view sequence between the first fields of view; wherein N is a positive integer greater than or equal to 2; Acquire a set of target objects that appear in every second field of view in the local field of view sequence; Matching the target object set with multiple standard object sets respectively; When the match is met, the key object indication rule of the matching standard object set is obtained; Based on the key object indication rule, determine the key objects and corresponding optimization categories from the target object set; When the optimization category is active optimization, the area of ​​the first visual field corresponding to the last occurrence of the key object in the first visual field in the trough of the appearance curve is taken as the first latest key area; When the optimization category is passive optimization, the area of ​​the overlapping fields of view between the target peaks corresponding to the first fields of view is taken as the first latest key area.

[0026] The field of view change sequence includes the first field of view of the user that has changed successively since the robot arm interacted with the user most recently; the first field of view refers to the user's eye field of view in the real environment, which can be determined by the robot arm detecting the user's eye position and orientation in combination with the standard eye field of view. The appearance degree of the robot arm in the first field of view refers to the ratio of the appearance area of ​​the robot arm in the first field of view to the total area of ​​the first field of view. The appearance degree curve drawn reflects the changing trend of the appearance degree of the robot arm in the user's first field of view.

[0027] When an emergency occurs during the interaction between the robot arm and the user, the user will hope that the robot arm can know the emergency and respond. The user will first check the robot arm preliminarily, and then check the robot arm several times in depth. During this period, the user will also check objects that can help the robot arm know the emergency. For example, if the user suddenly turns on the TV and wants to watch TV, but the robot arm may affect his vision when working next to it, the user will first check the robot arm preliminarily, hoping that it will avoid it, then check the TV, and then check the robot arm several times in depth, hoping that it will confirm and respond to avoid it, and will also check the TV. For another example, when the user is watching TV, if he wants the robot arm to accompany him, he will also check the robot arm preliminarily, hoping that it will approach, then check the TV, and then check the robot arm several times in depth, hoping that it will confirm and respond to approach, and will also check the TV. Therefore, after the first peak in the appearance curve, more than N target peaks higher than the first peak appear continuously. At this time, the local field of view sequence between the first field of view corresponding to the first peak of the target peak can reflect whether the user wants the robot arm to perform active optimization or passive optimization, for example: approach or avoid.

[0028] The target object set includes objects that appear in each second field of view, such as televisions, TV cabinets, snack carts, coffee tables, etc. A standard object set and corresponding key object indication rules are pre-set. The key object indication rules indicate how to determine from the target object set the objects that the robot arm needs to use as a basis for processing in order to adapt to emergencies, and also indicate the corresponding optimization categories. For example, if the standard object set includes televisions, TV cabinets, and snack carts, it means that the user not only wants to watch TV, but may also want to eat snacks in the middle of the process, and the robot arm can carry the snacks to the user, then the key object indication rule is to use the TV as the key object, and the optimization category is active optimization. For another example, if the standard object set includes televisions and TV cabinets, it means that the user only wants to watch TV, then the key object indication rule is to use the TV as the key object, and the optimization category is passive optimization.

[0029] When the optimization category is active optimization, the user wants the robot arm to get as close as possible without affecting the viewing of the key object. In this case, the appearance of the robot arm in the first field of view needs to be very low and the key object can be viewed. The user will also subjectively adjust the first field of view in the process of viewing the key object and the robot arm back and forth so that the robot arm knows the location area where the user wants the robot arm to be located. After continuous adjustment, the adjustment is finally successful. Therefore, the area of ​​the first field of view where the trough in the appearance curve corresponds to the last key object in the first field of view is taken as the first latest key area. When the optimization category is passive optimization, the user wants the robot arm to avoid as much as possible so as not to affect the viewing of the key object. The user will also subjectively adjust the first field of view in the process of viewing the key object and the robot arm back and forth so that the robot arm knows the location area where the user does not want the robot arm to be located. Therefore, the area of ​​the overlapping field of view between the target peaks corresponding to the first fields of view is taken as the first latest key area.

[0030] The embodiment of the present invention accurately identifies the first latest key area, ensures the effect of optimizing and replacing the second local trajectory entering the area, avoids excessive consumption of system resources due to over-identification, and avoids trajectory optimization that cannot enable the interaction between the robot arm and the user to flexibly adapt to emergencies due to under-identification, thereby greatly improving the applicability of the system.

[0031] In addition, a field of view change sequence is introduced, an appearance curve is drawn, and a trigger condition is set that more than N target peaks higher than the first peak appear continuously immediately after the first peak in the appearance curve. The timing of intercepting the first and last peaks of the target peak from the field of view change sequence corresponding to the local field of view sequence between the first fields of view is determined, and the target object set that is used as the basis for processing in order to make the robot arm adapt to emergencies is accurately determined. The standard object set and key object indication rules are used to quickly determine the key objects and the corresponding optimization categories. According to the different optimization categories, the appearance curve is used to determine the first and latest key areas in a targeted manner. This is progressive, which greatly improves the recognition efficiency and accuracy of the first and latest key areas. At the same time, it is also more intelligent.

[0032] In one embodiment, the optimizing and replacing processing of the second partial trajectory based on the first interaction situation between the robot arm and the user when the robot arm moves along the first partial trajectory includes: Performing feature description processing on the first interaction situation to obtain a first feature description vector; When the optimization category is active optimization, determining the active behavior corresponding to the first feature description vector from the active behavior library; Performing an optimization replacement process on the second local trajectory, so that the robot arm can exhibit active behavior in the first latest key area when moving along the second local trajectory after the optimization replacement process; When the optimization category is passive optimization, determining the passive behavior corresponding to the first feature description vector from the passive behavior library; The second local trajectory is optimized and replaced, so that the robot arm can show passive behavior in any area outside the first latest key area when moving along the second local trajectory after the optimization and replacement processing.

[0033] The working principle and beneficial effects of the above technical solution are: When the optimization category is active optimization, for example, the first interaction situation is that the user commands the robot arm to organize the snacks on the snack cart, which means that the user wants the robot arm to watch TV with him and be ready to carry and deliver snacks at any time. The first feature description vector after feature description processing corresponds to the active behavior of carrying and delivering snacks in the active behavior library, that is, the second local trajectory is optimized and replaced, so that the robot arm can show active behavior in the first latest key area when moving along the second local trajectory after the optimized replacement processing. The optimized replacement processing refers to optimizing the trajectory first and then replacing the original trajectory with the optimized trajectory. The robot's own trajectory planning ability can be used during optimization.

[0034] When the optimization category is passive optimization, for example, the first interaction situation is that the user has been watching TV and has not interacted with the robotic arm, which means that the user wants to watch TV with peace of mind. The passive behavior corresponding to the first feature description vector after feature description processing in the passive behavior library is to avoid entering the first latest key area, that is, to optimize and replace the second local trajectory so that the robotic arm can exhibit passive behavior in any area outside the first latest key area when moving along the second local trajectory after optimization and replacement. At this time, the second local trajectory can generally be directly replaced with the return trajectory of the first local trajectory.

[0035] The embodiment of the present invention performs feature description processing on the first interaction situation to obtain a first feature description vector, and determines the active behavior or passive behavior corresponding to the first feature description vector according to different optimization categories, and based on this, quickly performs targeted optimization and replacement processing on the second local trajectory, thereby greatly improving the efficiency of the optimization and replacement processing of the second local trajectory.

[0036] In one embodiment, drawing an appearance curve based on the appearance of the robot arm in each first field of view in the field of view change sequence comprises: Based on the curve drawing template, an appearance curve is drawn according to the appearance of the robot arm in each first field of view in the field of view change sequence; The horizontal axis of the template curve coordinate system in the curve drawing template is the order of field of view change, and the vertical axis of the template curve coordinate system is the magnitude of the appearance.

[0037] The curve drawing template contains a template curve coordinate system, which is a two-dimensional coordinate system. Its horizontal axis is the order of field of view changes, and its vertical axis is the magnitude of appearance. When drawing the appearance curve, first find multiple coordinate points in the template curve coordinate system according to the order of change of each first field of view and the corresponding magnitude of appearance, and connect each coordinate point in sequence to obtain the appearance curve. The curve drawing template is set in advance by the technician according to actual needs.

[0038] When the present application is actually used, a special problem scenario may arise: other users may newly enter the first and latest key area. At this time, if their stay time is long, it may also affect the user. However, as the user interacts with other users, the sudden situation caused by the original interaction between the robotic arm and the user may also change suddenly. The robotic arm needs to make further adaptive actions to avoid interfering with the user and other users.

[0039] Therefore, in order to overcome the above-mentioned special problem scenario, in one embodiment, after the second partial trajectory is optimized and replaced based on the first interaction situation of the robot arm interacting with the user when the robot arm moves along the first partial trajectory, the method further includes: After the robot arm finishes moving along the second local trajectory after the optimization replacement process, if other users stay in the first latest key area for more than a first preset time, identifying the second latest key area for interaction between the robot arm and the user and other users; Determine respectively from the second current future motion trajectory of the robot arm a third local trajectory before entering the second latest key area and a fourth local trajectory entering the second latest key area; Based on a second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory, the fourth partial trajectory is optimized and replaced.

[0040] The first preset duration is a threshold representing a longer stay, for example, the value is 20 seconds. The second latest key area is the interval where the robot arm needs to make the above-mentioned further adaptive actions and needs to perform local trajectory optimization. Accordingly, the third local trajectory before entering the second latest key area and the fourth local trajectory entering the second latest key area are respectively determined from the second current future motion trajectory of the robot arm. In order to avoid interfering with the user and other users after the robot arm makes the above-mentioned further adaptive actions, it is necessary to optimize and replace the fourth local trajectory based on the second interaction situation between the user and other users when the robot arm moves along the third local trajectory.

[0041] In the embodiment of the present invention, when other users stay in the first latest key area for a long time, the second latest key area for interaction between the robotic arm and the user and other users is identified, and based on the second interaction situation, the fourth local trajectory is optimized and replaced, so that the robotic arm can make further adaptive actions when encountering the above-mentioned special problem scenarios, so as to avoid interfering with the user and other users and overcome the above-mentioned special problem scenarios.

[0042] In one embodiment, the identifying the second latest key area for interaction between the robot arm and the user or other users includes: When the key object does not appear in the third visual field of the user and other users for more than a second preset time period, an overlapping area of ​​the third visual field of the user and other users that overlaps for more than a third preset time period is taken as the second latest key area.

[0043] The second preset time length is a threshold value representing a longer period of absence, for example, 100 seconds. The third preset time length is a threshold value representing a longer period of overlap, for example, 15 seconds. When a key object does not appear in the third field of view of the user and other users for more than the second preset time length, the area of ​​the overlapping field of view of the user and other users that overlaps for more than the third preset time length is taken as the second latest key area. For example, since other users entered the first latest key area, the user and the others have not watched TV for a long time, which means that they want to jointly create a new event, and the event will occur in the area of ​​the overlapping field of view where the two overlap for a long time, and it will be taken as the second latest key area.

[0044] The embodiment of the present invention accurately identifies the second latest key area, and also ensures the effect brought by the optimization replacement processing of the fourth local trajectory entering the second latest key area.

[0045] In one embodiment, the optimizing and replacing processing of the fourth partial trajectory based on the second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory includes: Performing feature description processing on the second interaction situation to obtain a second feature description vector; Determining the interaction triggering degree corresponding to the second feature description vector from the interaction triggering degree library; When the interaction trigger degree exceeds the trigger degree threshold, the robot arm is planned to continuously dock with the key object; The fourth local trajectory is optimized and replaced so that the robot arm can exhibit docking behavior in the second latest key area when moving along the fourth local trajectory after the optimization and replacement processing.

[0046] The second interaction situation will reflect whether the user and other users want the robot arm to participate in the interaction process between the two. The second interaction situation feature description is processed into a second feature description vector, and the corresponding interaction trigger degree is determined from the interaction trigger degree library. The interaction trigger degree reflects the degree to which the user and other users want the robot arm to participate in the interaction process between the two. For example, the second interaction situation is that the user and other users use a tablet computer together, which means that other users do not want the user to continue watching TV, but want the user to accompany them to watch the tablet computer. In this case, the robot arm is not wanted to participate in the interaction process between the two, and the interaction trigger degree is 0. For another example, the second interaction situation is that the user or other users send a review and summary of the content played on TV to the robot arm, so that other users can quickly understand the content played on TV and continue to watch TV with the user. The interaction trigger degree is 10. The trigger degree threshold is a threshold value representing the degree to which the user and other users want the robot arm to participate in the interaction process between the two, for example, the value is 5. When the interaction trigger degree exceeds the trigger degree threshold, the robot arm is planned to continuously dock with the key object. The docking behavior refers to the behavior of docking with the key object, for example: communication access to the TV, analysis of its played content for review and summary, and optimization and replacement of the fourth local trajectory, so that the robot arm can show docking behavior in the second latest key area when moving along the fourth local trajectory after the optimization and replacement processing.

[0047] The embodiment of the present invention introduces an interaction trigger degree. When the interaction trigger degree exceeds the trigger degree threshold, the docking behavior of the robotic arm to continuously dock with the key object is planned, and the fourth local trajectory is optimized and replaced, so that the robotic arm can exhibit docking behavior in the second latest key area when moving along the fourth local trajectory after the optimized replacement process, which greatly improves the accuracy and efficiency of the optimized replacement process of the fourth local trajectory.

[0048] The embodiment of the present invention provides a robot arm motion trajectory optimization system, such as Figure 2 As shown, including: The latest key area recognition module 1 is used to recognize the first latest key area for interaction between the robot arm and the user; A local trajectory determination module 2, used to determine a first local trajectory before entering the first latest key area and a second local trajectory entering the first latest key area from a first current future motion trajectory of the robot arm; The trajectory optimization and replacement processing module 3 is used to optimize and replace the second local trajectory based on a first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory.

[0049] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. A processor executes the computer program to implement any of the above methods.

[0050] An embodiment of the present invention provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any one of the methods described above.

[0051] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for optimizing the motion trajectory of a robotic arm, characterized in that: include: Identifying a first and latest critical area for interaction between the robotic arm and the user; Determine respectively from a first current future motion trajectory of the robot arm a first local trajectory before entering a first latest critical area and a second local trajectory entering the first latest critical area; Based on a first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory, an optimization replacement process is performed on the second local trajectory.

2. The method for optimizing the motion trajectory of a robotic arm according to claim 1, characterized in that: The identifying the first latest key area for interaction between the robot arm and the user comprises: Obtain the user's field of view change sequence after the robot arm last interacted with the user; Based on the appearance of the robot arm in each first field of view in the field of view change sequence, an appearance curve is drawn; When more than N target peaks higher than the first peak appear successively after the first peak in the appearance curve, the first and last peaks of the target peak are intercepted from the field of view change sequence, and each of them corresponds to a local field of view sequence between the first fields of view; wherein N is a positive integer greater than or equal to 2; Acquire a set of target objects that appear in every second field of view in the local field of view sequence; Matching the target object set with multiple standard object sets respectively; When the match is met, the key object indication rule of the matching standard object set is obtained; Based on the key object indication rule, determine the key objects and corresponding optimization categories from the target object set; When the optimization category is active optimization, the area of ​​the first visual field corresponding to the last occurrence of the key object in the first visual field in the trough of the appearance curve is taken as the first latest key area; When the optimization category is passive optimization, the area of ​​the overlapping fields of view between the target peaks corresponding to the first fields of view is taken as the first latest key area.

3. The method for optimizing the motion trajectory of a robotic arm according to claim 2, characterized in that: The optimizing and replacing processing of the second partial trajectory based on the first interaction situation between the robot arm and the user when the robot arm moves along the first partial trajectory includes: Performing feature description processing on the first interaction situation to obtain a first feature description vector; When the optimization category is active optimization, determining the active behavior corresponding to the first feature description vector from the active behavior library; Performing an optimization replacement process on the second local trajectory, so that the robot arm can exhibit active behavior in the first latest key area when moving along the second local trajectory after the optimization replacement process; When the optimization category is passive optimization, determining the passive behavior corresponding to the first feature description vector from the passive behavior library; The second local trajectory is optimized and replaced, so that the robot arm can show passive behavior in any area outside the first latest key area when moving along the second local trajectory after the optimization and replacement processing.

4. The method for optimizing the motion trajectory of a robotic arm according to claim 2, wherein: The step of drawing an appearance curve based on the appearance of the robot arm in each first field of view in the field of view change sequence comprises: Based on the curve drawing template, an appearance curve is drawn according to the appearance of the robot arm in each first field of view in the field of view change sequence; The horizontal axis of the template curve coordinate system in the curve drawing template is the order of field of view change, and the vertical axis of the template curve coordinate system is the magnitude of the appearance.

5. The method for optimizing the motion trajectory of a robot arm as claimed in claim 2, characterized in that: After the second partial trajectory is optimized and replaced based on the first interaction situation between the robot arm and the user when the robot arm moves along the first partial trajectory, the method further includes: After the robot arm finishes moving along the second local trajectory after the optimization replacement process, if other users stay in the first latest key area for more than a first preset time, identifying the second latest key area for interaction between the robot arm and the user and other users; Determine respectively from the second current future motion trajectory of the robot arm a third local trajectory before entering the second latest key area and a fourth local trajectory entering the second latest key area; Based on a second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory, the fourth partial trajectory is optimized and replaced.

6. The method for optimizing the motion trajectory of a robot arm according to claim 5, characterized in that: The second latest key area for identifying the interaction between the robot arm and the user or other users includes: When the key object does not appear in the third visual field of the user and other users for more than a second preset time period, an overlapping area of ​​the third visual field of the user and other users that overlaps for more than a third preset time period is taken as the second latest key area.

7. The method for optimizing the motion trajectory of a robot arm according to claim 5, characterized in that: The optimizing and replacing processing of the fourth partial trajectory based on the second interaction situation of the user interacting with other users when the robot arm moves along the third partial trajectory includes: Performing feature description processing on the second interaction situation to obtain a second feature description vector; Determining the interaction triggering degree corresponding to the second feature description vector from the interaction triggering degree library; When the interaction trigger degree exceeds the trigger degree threshold, the robot arm is planned to continuously dock with the key object; The fourth local trajectory is optimized and replaced so that the robot arm can exhibit docking behavior in the second latest key area when moving along the fourth local trajectory after the optimization and replacement processing.

8. A robot arm motion trajectory optimization system, characterized in that: include: A latest key area recognition module, used to recognize the first latest key area for interaction between the robot arm and the user; A local trajectory determination module, used to determine a first local trajectory before entering a first latest key area and a second local trajectory entering the first latest key area from a first current future motion trajectory of the robot arm; The trajectory optimization and replacement processing module is used to optimize and replace the second local trajectory based on a first interaction situation between the robot arm and the user when the robot arm moves along the first local trajectory.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 7.

Images