Intelligent control method and system for robot operation in sandy environment

Abstract

The present invention provides an intelligent control method and system for robot operations in a sandy environment, including: collecting real-time feedback data of a working robot, constructing a real-time working posture and a real-time motion strategy of the working robot according to the real-time feedback data, analyzing sand flow characteristics of the sandy environment, correcting the real-time motion strategy of the working robot according to the sand flow characteristics, acquiring map information of the sandy environment, analyzing the sand flow direction of the sandy environment in combination with the sand flow characteristics, performing path planning for the working robot in combination with the working goal of the working robot, analyzing a number of easy-to-sink areas included in the working path of the working robot, adjusting the working robot to an anti-sinking posture before the working robot reaches the easy-to-sink area, using an adaptive control algorithm, adjusting the robot's posture and motion strategy in real time according to sensor feedback, dynamically adjusting the path during the operation process, and adapting to changes in the sandy surface.

Description

Technical Field

[0001] The present invention relates to the technical field of robot intelligent control, and in particular to an intelligent control method and system for robot operation in a sandy environment. Background Art

[0002] When talking about sandy soil environment, people can't help but think of the insurmountable life-forbidden areas such as deserts, sand dunes, sandbanks, and beaches, which are daunting. However, through scientific research on sandy soil environment, we can obtain the geological features, plant and animal species and distribution of sandy soil environment, and provide direct evidence to reveal the changes in the geological features, soil, vegetation, climate, and water system of sandy soil environment, which has far-reaching significance for studying the development process and laws of desertification in agriculture, animal husbandry, ethnic economy, and human activity areas around deserts.

[0003] At present, with the rapid development of science and technology, the relevant technologies of sandy soil environment scientific research have gradually matured, and some high-tech sandy soil environment scientific research equipment have emerged, including robots. Although current robots can collect data on sandy soil environments, the collection process requires close cooperation between staff and drones, and the data collection efficiency is low. In addition, robots often sink into the sand during work, and staff need to be prepared to find robots at any time.

[0004] Therefore, the present invention provides an intelligent control method and system for robot operations in a sandy environment. Summary of the invention

[0005] The present invention discloses an intelligent control method and system for robot operation in a sandy soil environment. The method adopts an adaptive control algorithm to adjust the robot's posture and motion strategy in real time according to sensor feedback, and dynamically adjusts the path during the operation to adapt to changes in the sandy soil surface.

[0006] The present invention provides an intelligent control method for robot operation in a sandy environment, comprising:

[0007] Step 1: Collecting real-time feedback data of the working robot, and constructing the real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data;

[0008] Step 2: analyzing the sand flow characteristics of the sand environment according to the real-time working posture corresponding to each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics;

[0009] Step 3: Obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot;

[0010] Step 4: Analyze several easy-to-sink areas included in the working path of the working robot, and adjust the working robot to an anti-sinking posture before the working robot reaches the easy-to-sink areas.

[0011] In one practicable manner,

[0012] The step 1 comprises:

[0013] Step 11: When the working robot is working in a sandy environment, real-time feedback data of the working robot is collected, and a plurality of working information of the working robot is constructed according to the real-time feedback data. According to the logical relationship between different working information, each working information is regarded as a leaf node to construct an information logic tree of the working robot;

[0014] Step 12: Perform time slot aggregation on each leaf node to obtain the key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain the feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in the corresponding feedback component to obtain the component composition structure of the working robot;

[0015] Step 13: Generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine several combined key features of the working robot in combination with the key feedback features corresponding to each feedback component;

[0016] Step 14: Use the real-time working posture and the combined key features to construct posture change information of the working robot, and use the posture change information to generate a real-time motion strategy of the working robot.

[0017] In one practicable manner,

[0018] The step 2 comprises:

[0019] Step 21: determining the standard working posture of the working robot at the corresponding feedback moment according to the real-time motion strategy, counting the posture difference direction and posture difference amount between the standard working posture and the real-time working posture corresponding to the same feedback moment, and establishing the posture difference vector corresponding to each feedback moment;

[0020] Step 22: Time-sorting the difference vectors to generate a difference vector queue, dynamically monitoring the difference vector queue, obtaining the position dynamic offset information corresponding to the working robot at different feedback moments, and constructing the sand flow characteristics of the sand environment according to the position dynamic offset information;

[0021] Step 23: generating interference features of the working robot according to the sand flow features, calling a plurality of anti-interference strategies according to the interference features, and respectively docking each of the anti-interference strategies with the real-time motion strategy to obtain a matching degree between each of the anti-interference strategies and the real-time motion strategy;

[0022] Step 24: Filter the target anti-interference strategy with the highest matching degree, and use the target anti-interference strategy to merge with the real-time motion strategy to generate a corrected motion strategy for the working robot.

[0023] In one practicable manner,

[0024] Also includes:

[0025] After the working robot completes the work, recording the working posture change information of the working robot in the sand environment;

[0026] The terrain information of the sand environment is constructed according to the working posture change information, and the map information is updated using the terrain information.

[0027] In one practicable manner,

[0028] Also includes:

[0029] Establishing the working stability of the working robot according to the dynamic position offset information, and correcting the real-time motion strategy when the working stability does not meet the specified conditions;

[0030] Otherwise, no modification is made to the real-time motion strategy.

[0031] In one practicable manner,

[0032] The step 3 comprises:

[0033] Step 31: establishing a three-dimensional map model of the sandy environment according to the map information of the sandy environment, determining the regional sand flow direction corresponding to each model area in the three-dimensional map model according to the sand flow characteristics, and simulating the regional sand flow direction in the three-dimensional map model to obtain the sand flow direction of the sandy environment;

[0034] Step 32: obtaining a working target of the working robot, determining a minimum path of the working robot in the sandy environment based on the working target, marking the minimum path in the three-dimensional map model, and obtaining an overlapping area between the minimum path and the three-dimensional map model;

[0035] Step 33: Obtain the infeasible area in the overlapped area, adjust the minimum path in the three-dimensional map model until the infeasible area is eliminated to generate an adjusted path, simulate the actual path corresponding to the working robot moving along the adjusted path according to the sand flow direction, and establish path planning parameters according to the degree of deviation between the adjusted path and the actual path;

[0036] Step 34: Obtain the first path end point corresponding to the adjusted path and the second path end point corresponding to the actual path, and re-plan the adjusted path using the path planning parameters until the first path end point and the second path end point coincide with each other, thereby generating a working path for the working robot.

[0037] In one practicable manner,

[0038] The step 4 comprises:

[0039] Step 41: Divide the sandy soil environment into a plurality of sandy soil areas, determine the sand pit characteristics corresponding to each of the sandy soil areas according to the sandy soil flow direction, obtain a plurality of sandy soil areas overlapping with the working path, and analyze the risk probability of the working robot falling into the sandy soil area in combination with the sand pit characteristics corresponding to each of the sandy soil areas;

[0040] Step 42: screening a first easy-to-sink area within a first preset risk probability range of the risk probability of sinking, obtaining a backup risk probability of sinking corresponding to an adjacent sandy area of ​​the first easy-to-sink area, retrieving a corresponding backup direction according to the backup risk probability of sinking, and setting a first anti-sinking posture according to the backup direction;

[0041] Step 43: screening a second easy-to-sink area within a second preset risk probability range of the risk probability of sinking, determining a sand support force threshold corresponding to the second easy-to-sink area according to the sand pit characteristics, adjusting the working speed and working balance of the working robot according to the sand support force threshold, and constructing a second anti-sinking posture of the working robot;

[0042] Step 44: When the working robot arrives in front of the first easy-to-sink area, the working robot is controlled to adjust to the first anti-sink posture; when the working robot arrives in front of the second easy-to-sink area, the working robot is controlled to adjust to the second anti-sink posture.

[0043] In one practicable manner,

[0044] Also includes:

[0045] Determine a plurality of non-sinkable areas in the sandy soil environment according to the sinking risk probability corresponding to each of the sandy soil areas;

[0046] When the working robot arrives in front of the non-trapped area, the working robot is controlled to work according to the current posture.

[0047] The present invention provides an intelligent control system for robot operation in a sandy soil environment, comprising:

[0048] A real-time analysis module, used to collect real-time feedback data of the working robot, and construct a real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data;

[0049] A strategy correction module, used for analyzing the sand flow characteristics of the sand environment according to the corresponding real-time working posture at each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics;

[0050] A path planning module is used to obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot;

[0051] The anti-trapping execution module is used to analyze a number of easy-to-trap areas included in the working path of the working robot, and adjust the working robot to an anti-trapping posture before the working robot reaches the easy-to-trap areas.

[0052] In one practicable manner,

[0053] The real-time analysis module comprises:

[0054] A first analysis unit is used for collecting real-time feedback data of the working robot when the working robot is working in a sandy environment, constructing a plurality of working information of the working robot according to the real-time feedback data, and constructing an information logic tree of the working robot by treating each working information as a leaf node according to a logical relationship between different working information;

[0055] A second analysis unit is used to perform time slot aggregation on each leaf node to obtain key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in a corresponding feedback component to obtain a component composition structure of the working robot;

[0056] A third analysis unit is used to generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine a number of combined key features of the working robot in combination with the key feedback features corresponding to each of the feedback components;

[0057] The fourth analysis unit is used to construct the posture change information of the working robot by using the real-time working posture and the combined key feature, and generate the real-time motion strategy of the working robot by using the posture change information.

[0058] The achievable beneficial effects of the above technical solution are: in order to intelligently control the robot while ensuring its normal operation, the real-time working posture and real-time motion strategy of the working robot are constructed according to the real-time feedback data of the working robot, and then the real-time motion strategy of the working robot is adjusted according to the sand flow characteristics of the sand environment. At the same time, corresponding path planning is carried out for the working robot according to the sand flow direction of the sand environment. When the working robot starts working, the easy-to-sink areas contained in its working path are analyzed, and the working robot is controlled to make anti-sinking preparations in advance before reaching the easy-to-sink areas. In this way, the working robot can adjust its own motion strategy and posture according to the changes and unevenness of the sand surface, and consider the fluidity and irregular shape of the sand during the working robot's work, avoid selecting areas that may cause the robot to sink, and improve the working efficiency of the working robot.

[0059] 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, claims, and drawings.

[0060] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] 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:

[0062] Figure 1 A schematic diagram of the working process of an intelligent control method for robot operation in a sandy environment in an embodiment of the present invention;

[0063] Figure 2 The figure is a schematic diagram of the composition of an intelligent control system for robot operation in a sandy environment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] 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.

[0065] Example 1

[0066] This embodiment provides a method and system for intelligently controlling robot operations in a sandy environment. Figure 1 As shown, including:

[0067] Step 1: Collecting real-time feedback data of the working robot, and constructing the real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data;

[0068] Step 2: analyzing the sand flow characteristics of the sand environment according to the real-time working posture corresponding to each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics;

[0069] Step 3: Obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot;

[0070] Step 4: Analyze several easy-to-sink areas included in the working path of the working robot, and adjust the working robot to an anti-sinking posture before the working robot reaches the easy-to-sink areas.

[0071] In this example, the real-time feedback data refers to data that the working robot feeds back its own working conditions in real time when working in a sandy environment;

[0072] In this example, the working posture refers to the appearance of the working robot when working;

[0073] In this example, the motion strategy represents the forward / backward strategy executed by the working robot when working;

[0074] In this example, the map information represents a map of a sandy soil environment;

[0075] In this example, the sand flow direction refers to the phenomenon that the sand flows in one direction due to uneven particle size or high porosity of the sand in the sand environment;

[0076] In this example, the purpose of route planning is to ensure that the working robot can work normally and to ensure that the working robot will not be damaged.

[0077] The working principle and beneficial effects of the above technical solution are as follows: in order to intelligently control the robot while ensuring its normal operation, the real-time working posture and real-time motion strategy of the working robot are constructed according to the real-time feedback data of the working robot, and then the real-time motion strategy of the working robot is adjusted according to the sand flow characteristics of the sand environment. At the same time, corresponding path planning is carried out for the working robot according to the sand flow direction of the sand environment. When the working robot starts working, the easy-to-sink areas contained in its working path are analyzed, and the working robot is controlled to make anti-sinking preparations in advance before reaching the easy-to-sink areas. In this way, the working robot can adjust its own motion strategy and posture according to the changes and unevenness of the sand surface, and consider the fluidity and irregular shape of the sand during the working robot's work, avoid selecting areas that may cause the robot to sink, and improve the working efficiency of the working robot.

[0078] Example 2

[0079] On the basis of Example 1, the intelligent control method for robot operation in a sandy environment, step 1, comprises:

[0080] Step 11: When the working robot is working in a sandy environment, real-time feedback data of the working robot is collected, and a plurality of working information of the working robot is constructed according to the real-time feedback data. According to the logical relationship between different working information, each working information is regarded as a leaf node to construct an information logic tree of the working robot;

[0081] Step 12: Perform time slot aggregation on each leaf node to obtain the key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain the feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in the corresponding feedback component to obtain the component composition structure of the working robot;

[0082] Step 13: Generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine several combined key features of the working robot in combination with the key feedback features corresponding to each feedback component;

[0083] Step 14: Use the real-time working posture and the combined key features to construct posture change information of the working robot, and use the posture change information to generate a real-time motion strategy of the working robot.

[0084] In this example, the information logic tree is a binary tree established based on the logical relationship between different work information of the working robot. In the information logic tree, each work information can be analyzed independently;

[0085] In this example, one leaf node corresponds to one piece of work information;

[0086] In this example, time slot aggregation means aggregating spectrum resources in leaf nodes. The function of time slot aggregation is to aggregate valid information in each leaf node;

[0087] In this instance, the key feedback features represent the salient features presented by the leaf nodes;

[0088] In this example, a feedback component may correspond to one or more leaf nodes;

[0089] In this example, the component composition structure and the information logic tree have a corresponding logical relationship;

[0090] In this example, the combined key features represent the prominent features presented by the combination of different components in the working robot.

[0091] The working principle and beneficial effects of the above technical solution are as follows: when the working robot is performing work in a sandy environment, the real-time feedback data of the working robot is split and summarized to obtain several work information, and then an information logic tree about all the work information is established, and then the key feedback features of each leaf node in the information logic tree are analyzed through time slot aggregation, and the feedback components of each leaf node are determined by tracing the source, so as to construct the component composition structure of the working robot, and construct the real-time working posture of the working robot through three-dimensional mapping, and determine the combined key features of the working robot in combination with the connection relationship between different feedback components, and finally use the real-time working posture and combined key features to construct the posture change information of the working robot, thereby generating the real-time motion strategy of the working robot. The above technical means can be used to analyze the work of the working robot at every moment in the sandy environment, and the effectiveness and accuracy of the obtained real-time working posture and real-time motion strategy are guaranteed.

[0092] Example 3

[0093] On the basis of Example 1, the intelligent control method for robot operation in a sandy environment, step 2, comprises:

[0094] Step 21: determining the standard working posture of the working robot at the corresponding feedback moment according to the real-time motion strategy, counting the posture difference direction and posture difference amount between the standard working posture and the real-time working posture corresponding to the same feedback moment, and establishing the posture difference vector corresponding to each feedback moment;

[0095] Step 22: Time-sorting the difference vectors to generate a difference vector queue, dynamically monitoring the difference vector queue, obtaining the position dynamic offset information corresponding to the working robot at different feedback moments, and constructing the sand flow characteristics of the sand environment according to the position dynamic offset information;

[0096] Step 23: generating interference features of the working robot according to the sand flow features, calling a plurality of anti-interference strategies according to the interference features, and respectively docking each of the anti-interference strategies with the real-time motion strategy to obtain a matching degree between each of the anti-interference strategies and the real-time motion strategy;

[0097] Step 24: Filter the target anti-interference strategy with the highest matching degree, and use the target anti-interference strategy to merge with the real-time motion strategy to generate a corrected motion strategy for the working robot.

[0098] In this example, the standard working posture refers to the working posture presented by the working robot when executing the real-time motion strategy without external interference;

[0099] In this example, the posture difference direction represents the deviation angle between the direction corresponding to the standard working posture and the direction corresponding to the real-time working posture;

[0100] In this example, the posture difference represents the deviation between the posture level corresponding to the standard working posture and the posture level corresponding to the real-time working posture;

[0101] In this example, the posture difference vector is a vector whose direction is consistent with the posture difference direction and whose modulus is consistent with the posture difference amount;

[0102] In this example, the dynamic offset information represents the information of the movement of the working robot caused by external interference at different times.

[0103] The working principle and beneficial effects of the above technical solution: in order to ensure that the working robot can be working, the dynamic offset information of the working robot at different feedback moments is analyzed according to the posture difference vector of the working robot at different feedback moments, thereby indirectly reflecting the sand flow characteristics of the sand environment and determining the interference characteristics of the working robot, so as to call the corresponding anti-interference strategy, and select the target anti-interference strategy according to the degree of matching between different anti-interference strategies and real-time motion strategies, so as to integrate the target anti-interference strategy with the real-time motion strategy to generate a corrected motion strategy for the working robot. The working robot can perform corresponding movements under the guidance of the corrected motion strategy, thereby achieving the purpose of resisting external interference.

[0104] Example 4

[0105] On the basis of Example 1, the intelligent control method for robot operation in a sandy environment further includes:

[0106] After the working robot completes the work, recording the working posture change information of the working robot in the sand environment;

[0107] The terrain information of the sand environment is constructed according to the working posture change information, and the map information is updated using the terrain information.

[0108] The working principle and beneficial effects of the above technical solution: equipped with a real-time updated map to dynamically adjust the path during the operation to adapt to changes in the sand surface.

[0109] Example 5

[0110] On the basis of Example 3, the intelligent control method for robot operation in a sandy environment further includes:

[0111] Establishing the working stability of the working robot according to the dynamic position offset information, and correcting the real-time motion strategy when the working stability does not meet the specified conditions;

[0112] Otherwise, no modification is made to the real-time motion strategy.

[0113] In this example, the prescribed condition is that the working robot can perform the corresponding work and will not roll over.

[0114] The working principle and beneficial effects of the above technical solution are as follows: by being equipped with a real-time updated map, the path can be dynamically adjusted during the operation to adapt to changes in the sand surface.

[0115] Example 6

[0116] On the basis of Example 1, the intelligent control method for robot operation in a sandy environment is characterized in that step 3 comprises:

[0117] Step 31: establishing a three-dimensional map model of the sandy environment according to the map information of the sandy environment, determining the regional sand flow direction corresponding to each model area in the three-dimensional map model according to the sand flow characteristics, and simulating the regional sand flow direction in the three-dimensional map model to obtain the sand flow direction of the sandy environment;

[0118] Step 32: obtaining a working target of the working robot, determining a minimum path of the working robot in the sandy environment based on the working target, marking the minimum path in the three-dimensional map model, and obtaining an overlapping area between the minimum path and the three-dimensional map model;

[0119] Step 33: Obtain the infeasible area in the overlapped area, adjust the minimum path in the three-dimensional map model until the infeasible area is eliminated to generate an adjusted path, simulate the actual path corresponding to the working robot moving along the adjusted path according to the sand flow direction, and establish path planning parameters according to the degree of deviation between the adjusted path and the actual path;

[0120] Step 34: Obtain the first path end point corresponding to the adjusted path and the second path end point corresponding to the actual path, and re-plan the adjusted path using the path planning parameters until the first path end point and the second path end point coincide with each other, thereby generating a working path for the working robot.

[0121] In this example, the regional sand flow direction corresponding to each model area can be the same or different;

[0122] In this example, the minimum path means the working path with the shortest distance generated without considering the external environment;

[0123] In this example, the infeasible region refers to the region that the working robot cannot walk through;

[0124] In this example, the adjustment path represents a path along which the working robot can perform work;

[0125] In this example, the actual path refers to the path that the working robot takes when it deviates from the adjusted path due to interference from the external environment.

[0126] In this example, the first path end point indicates the end point of the adjusted path, and the second path end point indicates the end point of the actual path. The purpose of adjusting the first path end point and the second path end point to the same point is to ensure that the working robot can complete the work.

[0127] The working principle and beneficial effects of the above technical solution are as follows: the map information of the sandy environment is gathered to construct a three-dimensional map model, and then the regional sand flow direction of a model area is analyzed in the three-dimensional map model, and then the sand flow direction of the sandy environment is obtained by simulating the sand flow direction of each area, and then the minimum path is adjusted in the three-dimensional map model in combination with the working goal of the working robot, and the non-feasible area in the minimum path is eliminated, thereby determining the actual path of the working robot when executing the adjusted path, and adjusting the actual path and the end point of the adjusted path according to the sand flow direction, and finally replanning the path to obtain the working path of the working robot. In this way, the forward path of the working robot can be comprehensively analyzed to ensure that the working robot can perform normal work.

[0128] Example 7

[0129] On the basis of Example 1, the intelligent control method for robot operation in a sandy environment, step 4, comprises:

[0130] Step 41: Divide the sandy soil environment into a plurality of sandy soil areas, determine the sand pit characteristics corresponding to each of the sandy soil areas according to the sandy soil flow direction, obtain a plurality of sandy soil areas overlapping with the working path, and analyze the risk probability of the working robot falling into the sandy soil area in combination with the sand pit characteristics corresponding to each of the sandy soil areas;

[0131] Step 42: screening a first easy-to-sink area within a first preset risk probability range of the risk probability of sinking, obtaining a backup risk probability of sinking corresponding to an adjacent sandy area of ​​the first easy-to-sink area, retrieving a corresponding backup direction according to the backup risk probability of sinking, and setting a first anti-sinking posture according to the backup direction;

[0132] Step 43: screening a second easy-to-sink area within a second preset risk probability range of the risk probability of sinking, determining a sand support force threshold corresponding to the second easy-to-sink area according to the sand pit characteristics, adjusting the working speed and working balance of the working robot according to the sand support force threshold, and constructing a second anti-sinking posture of the working robot;

[0133] Step 44: When the working robot arrives in front of the first easy-to-sink area, the working robot is controlled to adjust to the first anti-sink posture; when the working robot arrives in front of the second easy-to-sink area, the working robot is controlled to adjust to the second anti-sink posture.

[0134] In this example, the first preset risk range indicates a range in which the probability of falling into risk is too high, and the range is [80%, 99%];

[0135] In this example, the second preset risk range represents a general range of the probability of risk occurrence, and its range is [70%, 79%];

[0136] In this example, the sand supporting capacity threshold value indicates the maximum supporting capacity of the sand in the second easily trapped area.

[0137] The working principle and beneficial effects of the above technical solution: In order to prevent the working robot from falling into the sand, the sand environment is first divided into several sand areas, and then the sand pit characteristics of each sand area are analyzed to analyze the risk probability of the working robot falling into the sand. Then, different trapping postures are set for areas with different probabilities. When the working robot reaches the corresponding easy-to-sink area, it is prepared to release the trap in advance. This can ensure that the working robot can complete the task smoothly and reduce the probability of its damage.

[0138] Example 8

[0139] On the basis of Example 7, the intelligent control method for robot operation in a sandy environment further includes:

[0140] Determine a plurality of non-sinkable areas in the sandy soil environment according to the sinking risk probability corresponding to each of the sandy soil areas;

[0141] When the working robot arrives in front of the non-trapped area, the working robot is controlled to work according to the current posture.

[0142] Example 9

[0143] This embodiment provides an intelligent control system for robot operation in a sandy environment. Figure 2 As shown, including:

[0144] A real-time analysis module, used to collect real-time feedback data of the working robot, and construct a real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data;

[0145] A strategy correction module, used for analyzing the sand flow characteristics of the sand environment according to the corresponding real-time working posture at each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics;

[0146] A path planning module is used to obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot;

[0147] The anti-trapping execution module is used to analyze a number of easy-to-trap areas included in the working path of the working robot, and adjust the working robot to an anti-trapping posture before the working robot reaches the easy-to-trap areas.

[0148] In this example, the real-time feedback data refers to data that the working robot feeds back its own working conditions in real time when working in a sandy environment;

[0149] In this example, the working posture refers to the appearance of the working robot when working;

[0150] In this example, the motion strategy represents the forward / backward strategy executed by the working robot when working;

[0151] In this example, the map information represents a map of a sandy soil environment;

[0152] In this example, the sand flow direction refers to the phenomenon that the sand flows in one direction due to uneven particle size or high porosity of the sand in the sand environment;

[0153] In this example, the purpose of route planning is to ensure that the working robot can work normally and to ensure that the working robot will not be damaged.

[0154] The working principle and beneficial effects of the above technical solution are as follows: in order to intelligently control the robot while ensuring its normal operation, the real-time working posture and real-time motion strategy of the working robot are constructed according to the real-time feedback data of the working robot, and then the real-time motion strategy of the working robot is adjusted according to the sand flow characteristics of the sand environment. At the same time, corresponding path planning is carried out for the working robot according to the sand flow direction of the sand environment. When the working robot starts working, the easy-to-sink areas contained in its working path are analyzed, and the working robot is controlled to make anti-sinking preparations in advance before reaching the easy-to-sink areas. In this way, the working robot can adjust its own motion strategy and posture according to the changes and unevenness of the sand surface, and consider the fluidity and irregular shape of the sand during the working robot's work, avoid selecting areas that may cause the robot to sink, and improve the working efficiency of the working robot.

[0155] Example 10

[0156] On the basis of Example 9, the intelligent control system for robot operation in a sandy environment, the real-time analysis module includes:

[0157] A first analysis unit is used for collecting real-time feedback data of the working robot when the working robot is working in a sandy environment, constructing a plurality of working information of the working robot according to the real-time feedback data, and constructing an information logic tree of the working robot by treating each working information as a leaf node according to a logical relationship between different working information;

[0158] A second analysis unit is used to perform time slot aggregation on each leaf node to obtain key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in a corresponding feedback component to obtain a component composition structure of the working robot;

[0159] A third analysis unit is used to generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine a number of combined key features of the working robot in combination with the key feedback features corresponding to each of the feedback components;

[0160] The fourth analysis unit is used to construct the posture change information of the working robot by using the real-time working posture and the combined key feature, and generate the real-time motion strategy of the working robot by using the posture change information.

[0161] In this example, the information logic tree is a binary tree established based on the logical relationship between different work information of the working robot. In the information logic tree, each work information can be analyzed independently;

[0162] In this example, one leaf node corresponds to one piece of work information;

[0163] In this example, time slot aggregation means aggregating spectrum resources in leaf nodes. The function of time slot aggregation is to aggregate valid information in each leaf node;

[0164] In this instance, the key feedback features represent the salient features presented by the leaf nodes;

[0165] In this example, a feedback component may correspond to one or more leaf nodes;

[0166] In this example, the component composition structure and the information logic tree have a corresponding logical relationship;

[0167] In this example, the combined key features represent the prominent features presented by the combination of different components in the working robot.

[0168] The working principle and beneficial effects of the above technical solution are as follows: when the working robot is performing work in a sandy environment, the real-time feedback data of the working robot is split and summarized to obtain several work information, and then an information logic tree about all the work information is established, and then the key feedback features of each leaf node in the information logic tree are analyzed through time slot aggregation, and the feedback components of each leaf node are determined by tracing the source, so as to construct the component composition structure of the working robot, and construct the real-time working posture of the working robot through three-dimensional mapping, and determine the combined key features of the working robot in combination with the connection relationship between different feedback components, and finally use the real-time working posture and combined key features to construct the posture change information of the working robot, thereby generating the real-time motion strategy of the working robot. The above technical means can be used to analyze the work of the working robot at every moment in the sandy environment, and the effectiveness and accuracy of the obtained real-time working posture and real-time motion strategy are guaranteed.

[0169] 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. An intelligent control method for robot operation in a sandy environment, characterized in that: include: Step 1: Collecting real-time feedback data of the working robot, and constructing the real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data; Step 2: analyzing the sand flow characteristics of the sand environment according to the real-time working posture corresponding to each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics; Step 3: Obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot; Step 4: analyzing a number of easy-to-sink areas included in the working path of the working robot, and adjusting the working robot to an anti-sinking posture before the working robot reaches the easy-to-sink area; The step 2 comprises: Step 21: determining the standard working posture of the working robot at the corresponding feedback moment according to the real-time motion strategy, counting the posture difference direction and posture difference amount between the standard working posture and the real-time working posture corresponding to the same feedback moment, and establishing the posture difference vector corresponding to each feedback moment; Step 22: Time-sorting the difference vectors to generate a difference vector queue, dynamically monitoring the difference vector queue, obtaining the position dynamic offset information corresponding to the working robot at different feedback moments, and constructing the sand flow characteristics of the sand environment according to the position dynamic offset information; Step 23: generating interference features of the working robot according to the sand flow features, calling a plurality of anti-interference strategies according to the interference features, and respectively docking each of the anti-interference strategies with the real-time motion strategy to obtain a matching degree between each of the anti-interference strategies and the real-time motion strategy; Step 24: Filter the target anti-interference strategy with the highest matching degree, and use the target anti-interference strategy to merge with the real-time motion strategy to generate a corrected motion strategy for the working robot; Also includes: Establishing the working stability of the working robot according to the dynamic position offset information, and correcting the real-time motion strategy when the working stability does not meet the specified conditions; Otherwise, no modification is made to the real-time motion strategy.

2. The intelligent control method for robot operation in a sandy environment as claimed in claim 1, characterized in that: The step 1 comprises: Step 11: When the working robot is working in a sandy environment, real-time feedback data of the working robot is collected, and a plurality of working information of the working robot is constructed according to the real-time feedback data. According to the logical relationship between different working information, each working information is regarded as a leaf node to construct an information logic tree of the working robot; Step 12: Perform time slot aggregation on each leaf node to obtain the key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain the feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in the corresponding feedback component to obtain the component composition structure of the working robot; Step 13: Generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine several combined key features of the working robot in combination with the key feedback features corresponding to each feedback component; Step 14: Use the real-time working posture and the combined key features to construct posture change information of the working robot, and use the posture change information to generate a real-time motion strategy of the working robot.

3. The intelligent control method for robot operation in a sandy environment as claimed in claim 1, characterized in that: Also includes: After the working robot completes the work, recording the working posture change information of the working robot in the sand environment; The terrain information of the sand environment is constructed according to the working posture change information, and the map information is updated using the terrain information.

4. The intelligent control method for robot operation in a sandy environment as claimed in claim 1, characterized in that: The step 3 comprises: Step 31: establishing a three-dimensional map model of the sandy environment according to the map information of the sandy environment, determining the regional sand flow direction corresponding to each model area in the three-dimensional map model according to the sand flow characteristics, and simulating the regional sand flow direction in the three-dimensional map model to obtain the sand flow direction of the sandy environment; Step 32: obtaining a working target of the working robot, determining a minimum path of the working robot in the sandy environment based on the working target, marking the minimum path in the three-dimensional map model, and obtaining an overlapping area between the minimum path and the three-dimensional map model; Step 33: Obtain the infeasible area in the overlapped area, adjust the minimum path in the three-dimensional map model until the infeasible area is eliminated to generate an adjusted path, simulate the actual path corresponding to the working robot moving along the adjusted path according to the sand flow direction, and establish path planning parameters according to the degree of deviation between the adjusted path and the actual path; Step 34: Obtain the first path end point corresponding to the adjusted path and the second path end point corresponding to the actual path, and re-plan the adjusted path using the path planning parameters until the first path end point and the second path end point coincide with each other, thereby generating a working path for the working robot.

5. The intelligent control method for robot operation in a sandy environment as claimed in claim 1, characterized in that: The step 4 comprises: Step 41: Divide the sandy soil environment into a plurality of sandy soil areas, determine the sand pit characteristics corresponding to each of the sandy soil areas according to the sandy soil flow direction, obtain a plurality of sandy soil areas overlapping with the working path, and analyze the risk probability of the working robot falling into the sandy soil area in combination with the sand pit characteristics corresponding to each of the sandy soil areas; Step 42: screening a first easy-to-sink area within a first preset risk probability range of the risk probability of sinking, obtaining a backup risk probability of sinking corresponding to an adjacent sandy area of ​​the first easy-to-sink area, retrieving a corresponding backup direction according to the backup risk probability of sinking, and setting a first anti-sinking posture according to the backup direction; Step 43: screening a second easy-to-sink area within a second preset risk probability range of the risk probability of sinking, determining a sand support force threshold corresponding to the second easy-to-sink area according to the sand pit characteristics, adjusting the working speed and working balance of the working robot according to the sand support force threshold, and constructing a second anti-sinking posture of the working robot; Step 44: When the working robot arrives in front of the first easy-to-sink area, the working robot is controlled to adjust to the first anti-sink posture; when the working robot arrives in front of the second easy-to-sink area, the working robot is controlled to adjust to the second anti-sink posture.

6. The intelligent control method for robot operation in a sandy environment as claimed in claim 5, characterized in that: Also included: Determine a plurality of non-sinkable areas in the sandy soil environment according to the sinking risk probability corresponding to each of the sandy soil areas; When the working robot arrives in front of the non-trapped area, the working robot is controlled to work according to the current posture.

7. An intelligent control system for robot operation in a sandy environment, characterized in that: include: A real-time analysis module, used to collect real-time feedback data of the working robot, and construct a real-time working posture and real-time motion strategy of the working robot according to the real-time feedback data; A strategy correction module, used for analyzing the sand flow characteristics of the sand environment according to the corresponding real-time working posture at each feedback moment, and correcting the real-time motion strategy of the working robot according to the sand flow characteristics; A path planning module is used to obtain map information of the sand environment, analyze the sand flow direction of the sand environment in combination with the sand flow characteristics, and plan a path for the working robot in combination with the working goal of the working robot; an anti-trapping execution module, configured to analyze a plurality of easy-to-trap areas included in a working path of the working robot, and adjust the working robot to an anti-trapping posture before the working robot reaches the easy-to-trap areas; The strategy correction module analyzes the sand flow characteristics of the sand environment according to the corresponding real-time working posture at each feedback moment, and the process of correcting the real-time motion strategy of the working robot according to the sand flow characteristics includes: Determine the standard working posture of the working robot at the corresponding feedback moment according to the real-time motion strategy, count the posture difference direction and posture difference amount between the standard working posture and the real-time working posture at the same feedback moment, and establish the posture difference vector corresponding to each feedback moment; The difference vectors are time-sorted to generate a difference vector queue, the difference vector queue is dynamically monitored to obtain the position dynamic offset information corresponding to the working robot at different feedback moments, and the sand flow characteristics of the sand environment are constructed according to the position dynamic offset information; Generate interference features of the working robot according to the sand flow features, call several anti-interference strategies according to the interference features, respectively connect each anti-interference strategy with the real-time motion strategy, and obtain the matching degree between each anti-interference strategy and the real-time motion strategy; Screening the target anti-interference strategy with the highest matching degree, and fusing the target anti-interference strategy with the real-time motion strategy to generate a corrected motion strategy for the working robot; Also includes: Establishing the working stability of the working robot according to the dynamic position offset information, and correcting the real-time motion strategy when the working stability does not meet the specified conditions; Otherwise, no modification is made to the real-time motion strategy.

8. The intelligent control system for robot operation in sandy soil environment as claimed in claim 7, characterized in that: The real-time analysis module comprises: A first analysis unit is used for collecting real-time feedback data of the working robot when the working robot is working in a sandy environment, constructing a plurality of working information of the working robot according to the real-time feedback data, and constructing an information logic tree of the working robot by treating each working information as a leaf node according to a logical relationship between different working information; A second analysis unit is used to perform time slot aggregation on each leaf node to obtain key feedback features contained in each leaf node, perform data tracing on each leaf node to obtain feedback components corresponding to each leaf node, combine the feedback components according to the information logic tree, and mark each key feedback feature in a corresponding feedback component to obtain a component composition structure of the working robot; A third analysis unit is used to generate the presentation features of the working robot according to the component composition structure, perform three-dimensional mapping on the presentation features to obtain the real-time working posture of the working robot, determine the connection relationship between different feedback components according to the component composition structure, and determine a number of combined key features of the working robot in combination with the key feedback features corresponding to each of the feedback components; The fourth analysis unit is used to construct the posture change information of the working robot by using the real-time working posture and the combined key feature, and generate the real-time motion strategy of the working robot by using the posture change information.

Images