Wheel arm redundant motion planning method based on artificial intelligence

Abstract

The invention discloses a wheel arm redundant motion planning method based on artificial intelligence, and belongs to the technical field of motion planning. Carrying out data integration calculation and analysis on the overall implementation of each time of the robot wheel arm redundant motion planning scheme according to a local planning monitoring processing set obtained through processing, and carrying out digital representation and classification on the single overall planning state of the robot wheel arm redundant motion planning scheme from the overall dimension; carrying out integrated analysis on single-time overall planning state data corresponding to different implementations of the robot wheel arm redundant motion planning schemes, and determining planning implementation reliable states corresponding to the robot wheel arm redundant motion planning schemes; dynamic optimization management is carried out on a robot wheel arm redundant motion planning scheme in a reliable state self-adaptive mode according to planning implementation; the robot wheel arm redundant motion planning method is used for solving the technical problems that in an existing scheme, the autonomous supervision analysis effect of robot wheel arm redundant motion planning scheme implementation is poor, and the self-adaptive planning optimization management effect is poor.

Description

Technical Field

[0001] The present invention relates to the technical field of motion planning, and in particular to a wheel arm redundant motion planning method based on artificial intelligence. Background Art

[0002] Wheel-arm redundant motion of a robot generally refers to the use of redundant degrees of freedom in robotic systems, especially those that combine a mobile platform (such as a wheeled base) and a robotic arm, to optimize motion paths, improve flexibility and obstacle avoidance capabilities; this design allows the robot to complete tasks in multiple ways, thereby providing more operating options and higher efficiency.

[0003] When implementing the existing robot wheel arm redundant motion planning scheme, it is impossible to conduct different aspects of supervision analysis and evaluation on the different planning implementation effects of the existing robot wheel arm redundant motion planning scheme, determine the planning implementation status corresponding to the existing robot wheel arm redundant motion planning scheme according to the evaluation results, and implement targeted optimization management. There is a poor autonomous supervision analysis effect and poor adaptive planning optimization management effect of the implementation of the robot wheel arm redundant motion planning scheme. Summary of the invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based wheel arm redundant motion planning method to solve the technical problems of poor autonomous supervision and analysis effect and poor adaptive planning optimization management effect in the existing robot wheel arm redundant motion planning scheme.

[0005] The purpose of the present invention can be achieved through the following technical solutions: The wheel arm redundant motion planning method based on artificial intelligence includes: Perform monitoring statistics on different aspects of planning processing for each implementation of the redundant motion planning scheme of the robot's wheel arm, perform local planning data processing on the monitoring statistical data of different aspects of planning processing and analyze and combine them to obtain a local planning monitoring processing set; Among them, when the local monitoring statistics and processing analysis of the effectiveness of the preliminary planning are carried out for each implementation of the redundant motion planning scheme of the robot wheel arm, all the planning angles JGi corresponding to different planning timestamps of different joints of the robot wheel arm are obtained, and all the planning angles corresponding to different joints are analyzed in turn through the planning angle identification function, and the effective identification of the angle planning JGBij corresponding to different planning angles is output; wherein, i is the different joints of the robot wheel arm, i=1, 2, 3, ..., n; n is a positive integer; j is the number of the planning action corresponding to different planning angles, j=1, 2, 3, ..., m; m is a positive integer; The angle planning validity flag contains a value of 0 or 1, indicating that the corresponding planning angle is valid or invalid; All effective identifications of angle planning obtained by corresponding processing of different joints of the robot wheel arm are sorted and combined respectively to obtain the angle planning processing sequence corresponding to different joints of the robot wheel arm; According to the local planning monitoring processing set obtained by processing, data integration calculation and analysis are performed on each overall implementation of the robot arm redundant motion planning scheme to obtain a single overall planning state data; The single overall planning status data corresponding to different implementations of the robot's wheel arm redundant motion planning schemes are integrated and analyzed to determine the planning implementation reliability status corresponding to the robot's wheel arm redundant motion planning scheme, and the robot's wheel arm redundant motion planning scheme is dynamically optimized and managed adaptively based on the planning implementation reliability status.

[0006] Preferably, the expression of the planning angle identification function is: ; In the formula, JG min ij, JG max ij are the minimum operating angle and maximum operating angle corresponding to different joints.

[0007] Preferably, the angle planning processing sequences corresponding to different joints of the robot wheel arm are traversed and analyzed in turn, and if all elements in the angle planning processing sequence are 0, the effective value of the angle planning associated with the corresponding wheel arm joint is set to 0; If there are non-zero elements in the angle planning processing sequence, the effective value of the angle planning associated with the corresponding wheel arm joint is set to N; N is the total number of non-zero elements in the angle planning processing sequence; The effective values ​​of the angle plans associated with all wheel-arm joints are sorted and combined to obtain an effective recognition sequence of the angle plans.

[0008] Preferably, when performing local monitoring statistics and processing analysis on the effectiveness of mid-term planning for each implementation of the redundant motion planning scheme of the robot's wheel arm, all obstacles identified and obtained during the implementation process of the redundant motion planning of the robot's wheel arm are monitored and obtained, as well as the obstacle types, identification discovery timestamps, identification processing end timestamps, and obstacle identification interval distances corresponding to all obstacles; Traverse and match different obstacle types with the pre-built obstacle type impact table to obtain the corresponding obstacle type impact identifier, and calculate the obstacle processing time corresponding to different obstacles based on the identification discovery timestamp and the identification processing end timestamp; The obstacle type impact identifier, obstacle processing time and obstacle identification interval distance obtained by the obstacle corresponding processing are sorted and combined to obtain a single obstacle identification processing sequence; The individual obstacle recognition processing sequences corresponding to different obstacles are sorted and combined in the order of discovery time to obtain an obstacle recognition processing sequence set.

[0009] Preferably, the individual obstacle identification processing sequences in the obstacle identification processing sequence set are sequentially processed by formula Calculate and obtain the local recognition reliability SKk corresponding to different obstacles; where k is the obstacle to be recognized, k=1, 2, 3, ..., p; p is a positive integer; GYk, TCk, LSk are the obstacle type impact identification, obstacle processing time, and obstacle recognition interval distance corresponding to different recognized obstacles; TC0k and LS0k are the obstacle processing standard time and obstacle recognition standard interval distance corresponding to different recognized obstacles; [ ] is a rounding function, which means obtaining the maximum integer that does not exceed a real number; The local recognition reliability obtained by processing different recognized obstacles is sorted and combined to obtain an effective processing sequence for obstacle planning; The angle planning effective identification sequence and the obstacle planning effective processing sequence constitute the local planning monitoring processing set.

[0010] Preferably, when performing data integration calculation on the overall implementation of the redundant motion planning scheme of the robot wheel arm according to the angle planning effective recognition sequence and the obstacle planning effective processing sequence in the local planning monitoring processing set, all the data in the angle planning effective recognition sequence and the obstacle planning effective processing sequence are calculated by the formula Calculate and obtain the single overall planning state value GZ of the redundant motion planning scheme of the robot wheel arm; where Nij is the effective value of different angle planning associated with different wheel arm joints; NJ1 is the total number of effective values ​​of angle planning associated with all wheel arm joints whose values ​​are not 0; a and b are the calculation weights of different data items, 0<a≤b, a+b=1; A is the standard value of single overall planning.

[0011] Preferably, if the single overall planning state value is 0, a single overall planning state normal label is generated and prompted; If the single overall planning status value is not 0, a single overall planning status abnormal label is generated and prompted; The single overall planning state value and the single overall planning state normal label or the single overall planning state abnormal label obtained through analysis constitute the single overall planning state data.

[0012] Preferably, all single overall planning state data are traversed and counted, and the total number of abnormal labels of the single overall planning state NY is calculated by the formula Calculate and obtain the overall impact value YL of the state abnormality corresponding to the redundant motion planning scheme of the robot wheel arm; NZ is the total number of all single overall planning state data; B is the overall standard impact rate of the state abnormality.

[0013] Preferably, the total number of angle planning valid values ​​NJ1 and the total number of local recognition reliability values ​​NJ2 of all wheel arm joints with non-zero values ​​are respectively calculated by the formula Calculate and obtain the state abnormal local influence value JZg corresponding to different local plans; where g=1,2; JZg is JZ1, JZ2, which are the local influence value of the first state abnormality and the local influence value of the second state abnormality respectively; NJg is NJ1, NJ2; NJZg is NJZ1, NJZ2, which are the total number of all planning angles corresponding to different joints of all robot wheel arm redundant motion planning schemes, and the total number of all obstacles identified and obtained by different implementation processes of all robot wheel arm redundant motion planning schemes; Cg is C1, C2, which are the local standard influence rate of the first state abnormality and the local standard influence rate of the second state abnormality respectively; The obtained state anomaly overall impact value, the first state anomaly local impact value, and the second state anomaly local impact value are sorted and combined to obtain a state anomaly impact integration sequence.

[0014] Preferably, the state abnormality affects the integration sequence and performs a traversal analysis; If the elements in the integration sequence affected by the abnormal state are all less than or equal to 0, it is indicated that the planning implementation state is reliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is maintained; If the abnormal state affects the existence of elements greater than 0 in the integrated sequence, it indicates that the planning implementation state is unreliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is dynamically adjusted.

[0015] Compared with the existing solutions, the present invention achieves the following beneficial effects: The present invention monitors and counts different aspects of planning processing for each implementation of the robot's wheel arm redundant motion planning scheme, performs local planning data processing on the monitoring statistical data of different aspects of planning processing and analyzes and combines them to obtain a local planning monitoring processing set, thereby improving different local autonomous supervision and analysis effects of the implementation of the robot's wheel arm redundant motion planning scheme.

[0016] The present invention performs data integration calculation and analysis on the overall implementation of each robot wheel arm redundant motion planning scheme according to the acquired local planning monitoring processing set, realizes the expansion and mining of local planning implementation processing data from different aspects in the early stage, and digitally represents and classifies the single overall planning state of the robot wheel arm redundant motion planning scheme from the overall dimension, thereby improving the overall autonomous supervision and analysis effect of the implementation of the robot wheel arm redundant motion planning scheme.

[0017] The present invention integrates and analyzes single overall planning state data corresponding to different implementations of the robot's wheel arm redundant motion planning schemes, determines the planning implementation reliability state corresponding to the robot's wheel arm redundant motion planning scheme, and adaptively dynamically optimizes and manages the robot's wheel arm redundant motion planning scheme according to the planning implementation reliability state, thereby improving the adaptive planning optimization management effect of the robot's wheel arm redundant motion planning scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below in conjunction with the accompanying drawings.

[0019] Figure 1 The present invention is a flowchart of the wheel arm redundant motion planning method based on artificial intelligence.

[0020] Figure 2 It is a schematic diagram of the structure of the redundant movement of the wheel arm in the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] like Figure 1-Figure 2 As shown, the present invention is a wheel arm redundant motion planning method based on artificial intelligence, comprising: Perform monitoring statistics on different aspects of planning processing for each implementation of the robot arm redundant motion planning scheme, perform local planning data processing and analysis and combination on the monitoring statistical data of different aspects of planning processing, and obtain a local planning monitoring processing set; including: When the local monitoring statistics and processing analysis of the effectiveness of the preliminary planning are carried out for each implementation of the redundant motion planning scheme of the robot wheel arm, all the planning angles JGi corresponding to different planning timestamps of different joints of the robot wheel arm are obtained. The unit of the planning timestamp is accurate to seconds, and all the planning angles corresponding to different joints are analyzed in turn through the planning angle identification function, and the angle planning valid identification JGBij corresponding to different planning angles is output; It should be noted that all planning angles corresponding to different joints of the robot wheel arm are automatically generated according to the existing robot wheel arm redundant motion planning scheme and the application requirements of the actual application scenario; Where i is the different joints of the robot wheel arm, i=1, 2, 3, ..., n; n is a positive integer; j is the number of the planned action corresponding to different planning angles, j=1, 2, 3, ..., m; m is a positive integer; The expression of planning angle identification function is: ; In the formula, JG min ij, JG max ij are the minimum operating angle and maximum operating angle corresponding to different joints, which are determined according to the operating design requirement data corresponding to different joints; The angle planning validity flag contains a value of 0 or 1, indicating that the corresponding planning angle is valid or invalid; All effective identifications of angle planning obtained by corresponding processing of different joints of the robot wheel arm are sorted and combined respectively to obtain the angle planning processing sequence corresponding to different joints of the robot wheel arm; In the embodiment of the present invention, by monitoring and counting the effectiveness of the preliminary planning and processing combination of each implementation of the redundant motion planning scheme of the robot wheel arm, the angle planning processing sequence corresponding to different joints of the robot wheel arm is obtained, which can not only realize the digital representation of the local angle planning state of different wheel arm joints, but also provide reliable local angle planning supervision data support for the subsequent angle planning effective state analysis; The angle planning processing sequences corresponding to different joints of the robot's wheel arm are traversed and analyzed in turn. If all the elements in the angle planning processing sequence are 0, the effective value of the angle planning associated with the corresponding wheel arm joint is set to 0; If there are non-zero elements in the angle planning processing sequence, the effective value of the angle planning associated with the corresponding wheel arm joint is set to N; N is the total number of non-zero elements in the angle planning processing sequence; The effective values ​​of the angle plans associated with all wheel arm joints are sorted and combined to obtain an effective recognition sequence of the angle plans; In an embodiment of the present invention, by traversing and analyzing the different angle planning processing sequences obtained in the early stage, and processing and sorting the angle planning effective values ​​of the traversal analysis results, an angle planning effective identification sequence is obtained, which can provide reliable angle planning effective analysis data support for the subsequent data analysis of the overall implementation status of each robot wheel arm redundant motion planning scheme.

[0023] When performing local monitoring statistics and processing analysis on the effectiveness of mid-term planning for each implementation of the redundant motion planning scheme of the robot's wheel arm, all obstacles identified and obtained during the implementation process of the redundant motion planning of the robot's wheel arm are monitored and obtained, as well as the obstacle types, identification discovery timestamps, identification processing end timestamps, and obstacle identification interval distances corresponding to all obstacles; Among them, the identification and acquisition of all obstacles can be determined according to the existing obstacle identification technology solutions, such as image recognition solutions, lidar recognition solutions, ultrasonic recognition solutions and multi-sensor fusion recognition solutions. The specific implementation steps are not described here. Obstacle types include but are not limited to static obstacle types, dynamic obstacle types, narrow space obstacle types, and soft obstacle types; Specifically, static obstacles are the most common type of obstacles, including walls, other machines and equipment, or fixed objects in the working environment. The positions of these obstacles are relatively fixed and can be avoided through pre-planning. Dynamic obstacle types: such as moving people, animals, or other robots; since their positions change over time, real-time perception combined with obstacle avoidance strategies are required to effectively avoid collisions; Small space obstacle type: Some work scenarios may require the robot arm to operate in a very confined space, such as internal inspection of pipelines or precision assembly processes; in this case, even the design of redundant degrees of freedom may face challenges; Soft obstacle type: Flexible objects such as cables and wires will not cause direct damage to the mechanical structure, but they can easily get entangled in the joints, affecting the motion trajectory and even causing failures; Traverse and match different obstacle types with the pre-built obstacle type impact table to obtain the corresponding obstacle type impact identifier, and calculate the obstacle processing time corresponding to different obstacles based on the identification discovery timestamp and the identification processing end timestamp; Among them, the obstacle type impact table pre-stores a number of sample obstacle types and their corresponding sample obstacle type impact identifiers. Different sample obstacle types are pre-associated with a corresponding sample obstacle type impact identifier. The sample obstacle type impact identifier is used to digitally represent the obstacle impact corresponding to the sample obstacle type. The specific value of the sample obstacle type impact identifier can be determined by the total number of historical occurrences of the corresponding sample obstacle type and the processing time. The obstacle type impact identifier, obstacle processing time and obstacle identification interval distance obtained by the obstacle corresponding processing are sorted and combined to obtain a single obstacle identification processing sequence; The individual obstacle recognition processing sequences corresponding to different obstacles are sorted and combined in the order of discovery time to obtain an obstacle recognition processing sequence set; In an embodiment of the present invention, by monitoring, counting and processing the effectiveness of mid-term planning for each implementation of the redundant motion planning scheme of the robot's wheel arm, an obstacle recognition processing sequence set corresponding to all obstacle planning processes is obtained, which can provide reliable processing data support for subsequent local recognition reliable state analysis corresponding to different obstacles.

[0024] The individual obstacle recognition processing sequences in the obstacle recognition processing sequence set are sequentially transformed into Calculate and obtain the local recognition reliability SKk corresponding to different obstacles; where k is the obstacle to be recognized, k=1, 2, 3, ..., p; p is a positive integer; GYk, TCk, LSk are the obstacle type impact identification, obstacle processing time, and obstacle recognition interval distance corresponding to different recognized obstacles; TC0k and LS0k are the obstacle processing standard time and obstacle recognition standard interval distance corresponding to different recognized obstacles, which are determined according to the design requirement data corresponding to the redundant motion of the robot wheel arm, or the previous obstacle test data; [ ] is a rounding function, which means obtaining the maximum integer that does not exceed a real number; The local recognition reliability obtained by processing different recognized obstacles is sorted and combined to obtain an effective processing sequence for obstacle planning; In the embodiment of the present invention, by calculating and sorting the data of the local identification reliable state of the planning processing data corresponding to different obstacles, an effective obstacle planning processing sequence corresponding to all obstacles is obtained, so as to realize the digital processing of the local identification reliable state corresponding to different obstacles, and at the same time, reliable local identification reliable supervision data support can be provided for the data analysis of each overall implementation state of the subsequent robot wheel arm redundant motion planning scheme; The effective recognition sequence of angle planning and the effective processing sequence of obstacle planning constitute the local planning monitoring processing set; In an embodiment of the present invention, by monitoring and statistics of different aspects of planning processing for each implementation of the robot's wheel arm redundant motion planning scheme, local planning data processing is performed on the monitoring statistical data of different aspects of planning processing and analyzed and combined to obtain a local planning monitoring processing set, thereby improving different local autonomous supervision and analysis effects of the implementation of the robot's wheel arm redundant motion planning scheme.

[0025] According to the local planning monitoring processing set obtained by processing, the data integration calculation and analysis are performed on each overall implementation of the robot arm redundant motion planning scheme to obtain the single overall planning state data; including: When the data integration calculation is performed on the overall implementation of the redundant motion planning scheme of the robot wheel arm according to the angle planning effective recognition sequence and the obstacle planning effective processing sequence in the local planning monitoring processing, all the data in the angle planning effective recognition sequence and the obstacle planning effective processing sequence are calculated by the formula Calculate and obtain the single overall planning state value GZ of the redundant motion planning scheme of the robot wheel arm; where Nij is the effective value of different angle planning associated with different wheel arm joints; NJ1 is the total number of effective values ​​of angle planning associated with all wheel arm joints whose values ​​are not 0; a and b are calculation weights of different data items, 0<a≤b, a+b=1, and the specific values ​​of a and b are not limited and can be customized according to the actual application requirements of the actual application scenario; A is the standard value of the single overall planning, which can be determined according to the design requirement data corresponding to the redundant motion of the robot wheel arm, or according to the previous motion test data; In the embodiment of the present invention, by integrating and calculating the local planning implementation processing data of different aspects in the early stage, a single overall planning state value of the robot wheel arm redundant motion planning scheme is obtained, thereby realizing the digital representation of the single overall planning state; If the single overall planning status value is 0, a single overall planning status normal label is generated and prompted; If the single overall planning status value is not 0, a single overall planning status abnormal label is generated and prompted; The single overall planning state value and the single overall planning state normal label or the single overall planning state abnormal label obtained through analysis constitute the single overall planning state data; In the embodiment of the present invention, data integration calculation and analysis are performed on each overall implementation of the robot wheel arm redundant motion planning scheme according to the local planning monitoring processing set obtained by processing, so as to realize the expansion and mining utilization of the local planning implementation processing data of different aspects in the early stage, and digitally represent and classify the single overall planning state of the robot wheel arm redundant motion planning scheme from the overall dimension, so as to improve the overall autonomous supervision and analysis effect of the implementation of the robot wheel arm redundant motion planning scheme; Integrate and analyze the single overall planning state data corresponding to different implementations of the robot wheel arm redundant motion planning scheme, determine the planning implementation reliability state corresponding to the robot wheel arm redundant motion planning scheme, and adaptively dynamically optimize the robot wheel arm redundant motion planning scheme according to the planning implementation reliability state; including: All single overall planning status data are traversed and counted, and the total number of abnormal labels of the single overall planning status NY is calculated through the formula Calculate and obtain the overall impact value YL of the abnormal state corresponding to the redundant motion planning scheme of the robot wheel arm; NZ is the total number of all single overall planning state data; B is the overall standard impact rate of the abnormal state, which can be determined according to the design requirement data corresponding to the redundant motion of the robot wheel arm, or according to the previous motion test data; It should be noted that the overall impact value of the abnormal state is used to calculate the overall impact of the abnormal state on the overall dimension of all implementation results of the redundant motion planning scheme of the robot arm; And, the total number of angle planning valid values ​​NJ1 and the total number of local recognition reliability NJ2 of non-zero values ​​associated with all wheel arm joints are respectively calculated by the formula Calculate and obtain the state abnormal local influence value JZg corresponding to different local plans; where g=1,2; JZg is JZ1 and JZ2, which are the first state abnormal local influence value and the second state abnormal local influence value respectively; NJg is NJ1 and NJ2; NJZg is NJZ1 and NJZ2, which are the total number of all planned angles of different joints corresponding to all robot wheel arm redundant motion planning schemes and the total number of all obstacles identified and obtained by different implementation processes corresponding to all robot wheel arm redundant motion planning schemes; Cg is C1 and C2, which are the first state abnormal local standard influence rate and the second state abnormal local standard influence rate respectively, which can be determined according to the design requirement data corresponding to the robot wheel arm redundant motion, or according to the previous motion test data; It should be noted that the first state abnormal local impact value and the second state abnormal local impact value are respectively used to calculate the data of the degree of local impact of the state abnormality for different local dimensions of all implementation results of the robot wheel arm redundant motion planning scheme; The overall impact value of the state anomaly, the local impact value of the first state anomaly, and the local impact value of the second state anomaly obtained by processing are sorted and combined to obtain a state anomaly impact integration sequence; In an embodiment of the present invention, by calculating and combining data on the degree of influence of corresponding state abnormalities in the overall dimensions and different local dimensions of all implementation results of the robot wheel arm redundant motion planning scheme, reliable multi-dimensional data support can be provided for the reliable state data analysis of the planning implementation corresponding to the robot wheel arm redundant motion planning scheme.

[0026] Conduct traversal analysis on the integration sequence affected by abnormal status; If the elements in the integration sequence affected by the abnormal state are all less than or equal to 0, it is indicated that the planning implementation state is reliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is maintained; If the state abnormality affects the existence of elements greater than 0 in the integrated sequence, it indicates that the planning implementation state is unreliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is dynamically adjusted; Among them, the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is dynamically adjusted. Specifically, the planning and design optimization adjustment of the corresponding local dimension and the planning and design optimization adjustment of the corresponding overall dimension can be implemented according to the elements greater than 0.

[0027] In an embodiment of the present invention, single overall planning status data corresponding to different implementations of the robot's wheel arm redundant motion planning scheme are integrated and analyzed to determine the planning implementation reliability status corresponding to the robot's wheel arm redundant motion planning scheme, and the robot's wheel arm redundant motion planning scheme is adaptively dynamically optimized and managed based on the planning implementation reliability status, thereby improving the adaptive planning optimization management effect of the robot's wheel arm redundant motion planning scheme.

[0028] In addition, the formulas involved in the above are all dimensionless and numerical calculations. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and simulating it with simulation software.

[0029] In the several embodiments provided by the present invention, it should be understood that the disclosed method can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0030] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0031] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0032] It is obvious to a person skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A wheel arm redundant motion planning method based on artificial intelligence, characterized in that: include: Perform monitoring statistics on different aspects of planning processing for each implementation of the redundant motion planning scheme of the robot's wheel arm, perform local planning data processing on the monitoring statistical data of different aspects of planning processing and analyze and combine them to obtain a local planning monitoring processing set; Among them, when the local monitoring statistics and processing analysis of the effectiveness of the preliminary planning are carried out for each implementation of the redundant motion planning scheme of the robot wheel arm, all the planning angles JGi corresponding to different planning timestamps of different joints of the robot wheel arm are obtained, and all the planning angles corresponding to different joints are analyzed in turn through the planning angle identification function, and the effective identification of the angle planning JGBij corresponding to different planning angles is output; wherein, i is the different joints of the robot wheel arm, i=1, 2, 3, ..., n; n is a positive integer; j is the number of the planning action corresponding to different planning angles, j=1, 2, 3, ..., m; m is a positive integer; The angle planning validity flag contains a value of 0 or 1, indicating that the corresponding planning angle is valid or invalid; All effective identifications of angle planning obtained by corresponding processing of different joints of the robot wheel arm are sorted and combined respectively to obtain the angle planning processing sequence corresponding to different joints of the robot wheel arm; According to the local planning monitoring processing set obtained by processing, data integration calculation and analysis are performed on each overall implementation of the robot arm redundant motion planning scheme to obtain a single overall planning state data; The single overall planning status data corresponding to different implementations of the robot's wheel arm redundant motion planning schemes are integrated and analyzed to determine the planning implementation reliability status corresponding to the robot's wheel arm redundant motion planning scheme, and the robot's wheel arm redundant motion planning scheme is dynamically optimized and managed adaptively based on the planning implementation reliability status.

2. The wheel arm redundant motion planning method based on artificial intelligence according to claim 1 is characterized in that: The expression of planning angle identification function is: ; In the formula, JG min ij, JG max ij are the minimum operating angle and maximum operating angle corresponding to different joints.

3. The wheel arm redundant motion planning method based on artificial intelligence according to claim 2 is characterized in that: The angle planning processing sequences corresponding to different joints of the robot's wheel arm are traversed and analyzed in turn. If all the elements in the angle planning processing sequence are 0, the effective value of the angle planning associated with the corresponding wheel arm joint is set to 0; If there are non-zero elements in the angle planning processing sequence, the effective value of the angle planning associated with the corresponding wheel arm joint is set to N; N is the total number of non-zero elements in the angle planning processing sequence; The effective values ​​of the angle plans associated with all wheel-arm joints are sorted and combined to obtain an effective recognition sequence of the angle plans.

4. The wheel arm redundant motion planning method based on artificial intelligence according to claim 3 is characterized in that: When performing local monitoring statistics and processing analysis on the effectiveness of mid-term planning for each implementation of the redundant motion planning scheme of the robot's wheel arm, all obstacles identified and obtained during the implementation process of the redundant motion planning of the robot's wheel arm are monitored and obtained, as well as the obstacle types, identification discovery timestamps, identification processing end timestamps, and obstacle identification interval distances corresponding to all obstacles; Traverse and match different obstacle types with the pre-built obstacle type impact table to obtain the corresponding obstacle type impact identifier, and calculate the obstacle processing time corresponding to different obstacles based on the identification discovery timestamp and the identification processing end timestamp; The obstacle type impact identifier, obstacle processing time and obstacle identification interval distance obtained by the obstacle corresponding processing are sorted and combined to obtain a single obstacle identification processing sequence; The individual obstacle recognition processing sequences corresponding to different obstacles are sorted and combined in the order of discovery time to obtain an obstacle recognition processing sequence set.

5. The wheel arm redundant motion planning method based on artificial intelligence according to claim 4 is characterized in that: The individual obstacle recognition processing sequences in the obstacle recognition processing sequence set are sequentially transformed into Calculate and obtain the local recognition reliability SKk corresponding to different obstacles; where k is the obstacle to be recognized, k=1, 2, 3, ..., p; p is a positive integer; GYk, TCk, LSk are the obstacle type impact identification, obstacle processing time, and obstacle recognition interval distance corresponding to different recognized obstacles; TC0k and LS0k are the obstacle processing standard time and obstacle recognition standard interval distance corresponding to different recognized obstacles; [ ] is a rounding function, which means obtaining the maximum integer that does not exceed a real number; The local recognition reliability obtained by processing different recognized obstacles is sorted and combined to obtain an effective processing sequence for obstacle planning; The angle planning effective identification sequence and the obstacle planning effective processing sequence constitute the local planning monitoring processing set.

6. The wheel arm redundant motion planning method based on artificial intelligence according to claim 5 is characterized in that: When the data integration calculation is performed on the overall implementation of the redundant motion planning scheme of the robot wheel arm according to the angle planning effective recognition sequence and the obstacle planning effective processing sequence in the local planning monitoring processing, all the data in the angle planning effective recognition sequence and the obstacle planning effective processing sequence are calculated by the formula Calculate and obtain the single overall planning state value GZ of the redundant motion planning scheme of the robot wheel arm; where Nij is the effective value of different angle planning associated with different wheel arm joints; NJ1 is the total number of effective values ​​of angle planning associated with all wheel arm joints whose values ​​are not 0; a and b are the calculation weights of different data items, 0<a≤b, a+b=1; A is the standard value of single overall planning.

7. The method for redundant motion planning of wheel arms based on artificial intelligence according to claim 6, characterized in that: If the single overall planning status value is 0, a single overall planning status normal label is generated and prompted; If the single overall planning status value is not 0, a single overall planning status abnormal label is generated and prompted; The single overall planning state value and the single overall planning state normal label or the single overall planning state abnormal label obtained through analysis constitute the single overall planning state data.

8. The wheel arm redundant motion planning method based on artificial intelligence according to claim 7 is characterized in that: All single overall planning status data are traversed and counted, and the total number of abnormal labels of the single overall planning status NY is calculated through the formula Calculate and obtain the overall impact value YL of the state abnormality corresponding to the redundant motion planning scheme of the robot wheel arm; NZ is the total number of all single overall planning state data; B is the overall standard impact rate of the state abnormality.

9. The method for redundant motion planning of wheel arms based on artificial intelligence according to claim 8, characterized in that: The total number of angle planning valid values ​​NJ1 and the total number of local recognition reliability NJ2 of all wheel arm joints with non-zero values ​​are respectively calculated by the formula Calculate and obtain the state abnormal local impact value JZg corresponding to different local plans; where g=1, 2; JZg is JZ1, JZ2, which are the first state abnormal local impact value and the second state abnormal local impact value respectively; NJg is NJ1 and NJ2; NJZg is NJZ1 and NJZ2, which are the total number of all planning angles corresponding to different joints of all robot wheel arm redundant motion planning schemes and the total number of all obstacles identified and obtained by different implementation processes of all robot wheel arm redundant motion planning schemes; Cg is C1 and C2, which are the local standard impact rate of abnormality in the first state and the local standard impact rate of abnormality in the second state, respectively; The obtained state anomaly overall impact value, the first state anomaly local impact value, and the second state anomaly local impact value are sorted and combined to obtain a state anomaly impact integration sequence.

10. The wheel arm redundant motion planning method based on artificial intelligence according to claim 9 is characterized in that: Conduct traversal analysis on the integration sequence affected by abnormal status; If the elements in the integration sequence affected by the abnormal state are all less than or equal to 0, it is indicated that the planning implementation state is reliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is maintained; If the abnormal state affects the existence of elements greater than 0 in the integrated sequence, it indicates that the planning implementation state is unreliable, and the subsequent implementation of the existing robot wheel arm redundant motion planning scheme is dynamically adjusted.

Images