A high-efficiency path planning and obstacle avoidance optimization method and system special for robots

Abstract

The application is suitable for the technical field of underwater robot path planning and intelligent obstacle avoidance control, and provides a robot-specific efficient path planning and obstacle avoidance optimization method and system.The method comprises the following steps: when it is determined that the angle deviation prediction of the target robot to the narrow area to be passed is less than the upper limit value of the angle deviation, the current environment and the traffic environment dynamic influence parameters of the narrow area to be passed are acquired, and a historical cleaning operation database is called.The application realizes the dynamic perception and adjustment of the underwater robot to the slight performance difference and the environmental disturbance influence in the process of rapidly passing through the narrow area for the first time by constructing an angle deviation prediction correction mechanism based on the response lag characteristics of the propulsion motor.

Description

Technical Field

[0001] This invention belongs to the field of underwater robot path planning and intelligent obstacle avoidance control technology, and particularly relates to a high-efficiency path planning and obstacle avoidance optimization method and system for robots. Background Technology

[0002] In existing underwater robot path planning and obstacle avoidance control technologies, feasibility assessments for traversing areas are typically based on statically constructed angle offset prediction models. These models largely rely on ideal propulsion behavior parameters and environmental response characteristics, making it difficult to fully consider the dynamic response differences of propulsion motors under various operating conditions. This is especially true when the robot needs to quickly traverse narrow structural areas, where the correlation between thrust response and heading deviation is difficult to accurately capture. Therefore, when complex factors such as fluid disturbances, changes in water viscosity, or structural vortex interference exist in the robot's operating environment, existing models may exhibit prediction biases, leading to misjudgments in the path planning system and impacting operational safety and continuity.

[0003] In actual underwater operations, robots often need to traverse narrow passages at high speeds to efficiently cover or avoid obstacles. In such cases, the control system typically sends acceleration commands to the propulsion motors to achieve rapid passage. However, there is a response lag between the issuance of the command and the actual output of the propulsion motor reaching the target thrust. This lag is significantly volatile, influenced by both the robot's own performance and environmental disturbances. If the thrust instability caused by this lag is not accurately identified and addressed by the system, it can lead to additional angular deviations, causing the robot to deviate from its preset path or even collide with obstacles. Current technology lacks quantitative assessment and dynamic correction methods for the risk of angular deviation under such dynamic changes, resulting in insufficient obstacle avoidance capabilities in boundary conditions and potential misjudgments and safety hazards. Summary of the Invention

[0004] The purpose of this invention is to provide a highly efficient path planning and obstacle avoidance optimization method and system specifically for robots, aiming to solve the problems mentioned in the background art.

[0005] This invention is implemented as follows: a highly efficient path planning and obstacle avoidance optimization method specifically for robots, the method comprising:

[0006] When it is determined that the predicted angular offset of the target robot in the narrow area to be traversed is less than the upper limit of the angular offset, the dynamic influence parameters of its current environment and the passage environment of the narrow area to be traversed are obtained, and the historical cleaning operation database is called.

[0007] Based on the database, reference cleaning samples of other reference robots that are the same type as the target robot, have the same background parameters of the propulsion motor, and whose historical environment and the dynamic influence parameters of the passage environment corresponding to the narrow area to be crossed in the past are respectively consistent.

[0008] By analyzing clean samples, we statistically analyzed the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from the historical environment.

[0009] Calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and set the reference clean samples with deviation exceeding the preset range as abnormal samples;

[0010] Determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, calculate an adjustment factor based on the deviation magnitude of all abnormal samples, apply the adjustment factor to the angle offset prediction, obtain the corrected angle offset prediction, and determine whether the target robot is allowed to pass through the narrow area to be traversed.

[0011] As a further limitation of the technical solution of the embodiment of the present invention, the dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighted processing or functional relationship calculation of the two, which are used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area.

[0012] As a further limitation of the technical solution of this invention, the steps of analyzing and referring to clean samples and statistically analyzing the response lag time from the control signal triggering to the actual generation of the required thrust by the propulsion motor during the process of entering the narrow area to be traversed from its historical environment include:

[0013] Each reference clean sample is analyzed sequentially to extract the timestamp of the control signal issued by the reference robot control system when it instructs the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, as well as the timestamp of the propulsion motor actually reaching the required thrust for crossing the historical area to be crossed.

[0014] Calculate the time difference between the two timestamps as the response lag time for the reference clean sample.

[0015] As a further limitation of the technical solution of this embodiment of the invention, the step of calculating the deviation of the response lag time corresponding to each reference clean sample from the preset standard lag time, and setting the reference clean sample whose deviation exceeds the preset range as an abnormal sample includes:

[0016] Obtain a preset standard lag time, which is the reference response time corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within a preset low threshold.

[0017] Calculate the deviation of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time.

[0018] Reference clean samples whose deviation exceeds a preset deviation threshold are set as abnormal samples.

[0019] As a further limitation of the technical solution of this embodiment of the invention, the step of determining whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio, and if so, calculating an adjustment factor based on the deviation magnitude corresponding to all abnormal samples, applying the adjustment factor to the angle offset prediction, obtaining the corrected angle offset prediction, and determining whether the target robot is allowed to pass through the narrow area to be traversed based on this includes:

[0020] Calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and determine whether it exceeds a preset ratio threshold;

[0021] If the deviation exceeds the limit, obtain the deviation of the response lag time of all abnormal samples relative to the preset standard lag time, and set the average value of all deviations as the adjustment factor;

[0022] The preset angle offset prediction correction function is called to correct the initial angle offset prediction based on the adjustment factor, and the corrected angle offset prediction is obtained.

[0023] The corrected angle offset prediction is compared with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed. If it is not less than the upper limit of the angle offset, it is determined that it cannot pass through.

[0024] As a further limitation of the technical solution of this embodiment of the invention, the angle offset prediction correction function is:

[0025] ;

[0026] in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time;

[0027] It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

[0028] A high-efficiency path planning and obstacle avoidance optimization system for robots, the system comprising: a data acquisition module, a reference sample acquisition module, a lag duration determination module, an abnormal sample determination module, and a prediction correction module, wherein:

[0029] The data acquisition module is used to acquire the current environment of the target robot and the dynamic influence parameters of the passage environment of the narrow area to be traversed when it is determined that the predicted angular offset of the target robot in the narrow area to be traversed is less than the upper limit of the angular offset, and to call the historical cleaning operation database.

[0030] The reference sample acquisition module is used to filter out reference cleaning samples of other reference robots that are consistent with the target robot type, have the same propulsion motor background parameters, and whose historical environment and the corresponding dynamic influence parameters of the passage environment of the narrow area to be traversed are consistent, based on the database. The dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighting the two or by calculating the functional relationship, which are used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area.

[0031] The lag time determination module is used to analyze the reference clean sample and calculate the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from its historical environment.

[0032] The abnormal sample determination module is used to calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and to set the reference clean sample whose deviation exceeds the preset range as an abnormal sample.

[0033] The prediction correction module is used to determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, it calculates an adjustment factor based on the deviation magnitude of all abnormal samples, applies the adjustment factor to the angle offset prediction, obtains the corrected angle offset prediction, and determines whether the target robot is allowed to pass through the narrow area to be traversed.

[0034] As a further limitation of the technical solution of this embodiment of the invention, the delay duration determination module specifically includes:

[0035] The timestamp acquisition unit is used to sequentially parse each reference clean sample, extract the timestamp of the control signal issued by the reference robot control system when instructing the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, and the timestamp of the propulsion motor actually reaching the required thrust to cross the historical area to be crossed.

[0036] The time difference calculation unit is used to calculate the time difference between two timestamps, which serves as the response lag time of the reference clean sample.

[0037] As a further limitation of the technical solution of this embodiment of the invention, the abnormal sample determination module specifically includes:

[0038] The standard duration acquisition unit is used to acquire a preset standard lag duration, which is the reference response duration corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within a preset low threshold.

[0039] The deviation magnitude calculation unit is used to calculate the deviation magnitude of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation magnitude is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time.

[0040] The abnormal sample setting unit is used to set reference clean samples whose deviation exceeds a preset deviation threshold as abnormal samples.

[0041] As a further limitation of the technical solution of this embodiment of the invention, the prediction correction module specifically includes:

[0042] The ratio judgment unit is used to calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and to determine whether it exceeds the preset ratio threshold.

[0043] The adjustment factor determination unit is used to obtain the deviation of the response lag time of all abnormal samples from the preset standard lag time if the deviation exceeds the limit, and set the average value of all deviations as the adjustment factor.

[0044] The prediction correction unit is used to call the preset angle offset prediction correction function to correct the initial angle offset prediction based on the adjustment factor, so as to obtain the corrected angle offset prediction.

[0045] The corrected prediction unit is used to compare the corrected angle offset prediction with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed; if it is not less than the upper limit, it is determined that it cannot pass through.

[0046] The angle offset prediction correction function is:

[0047] ;

[0048] in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time;

[0049] It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] This invention, by constructing an angle offset prediction and correction mechanism based on the response lag characteristics of the propulsion motor, achieves for the first time the dynamic perception and adjustment of the impact of minute performance differences and environmental disturbances on underwater robots during rapid passage through narrow areas. Compared with the existing technology that uses static prediction models or fixed threshold judgments, this invention introduces "response lag time deviation" as a quantitative indicator reflecting the coupling state between propulsion motor performance and the passage environment, and transforms it into a correction factor for the angle offset prediction, effectively enhancing the adaptability of the prediction model under non-ideal conditions. This mechanism can promptly identify risky conditions that may cause additional offsets and dynamically determine whether the robot is allowed to perform the crossing operation, thereby improving the judgment accuracy and robustness of the path planning and obstacle avoidance system in complex underwater environments, and has good engineering practical value and promotion prospects. Attached Figure Description

[0052] Figure 1 A flowchart of the method provided in the embodiments of the present invention;

[0053] Figure 2 This is a flowchart illustrating the method for determining the response hysteresis time corresponding to a reference clean sample in the embodiments of the present invention;

[0054] Figure 3 This is a flowchart illustrating the method for determining abnormal samples provided in this embodiment of the invention;

[0055] Figure 4 This is a flowchart illustrating the correction of the initial angle offset prediction in the method provided in this embodiment of the invention;

[0056] Figure 5 Application architecture diagram of the system provided in the embodiments of the present invention;

[0057] Figure 6 This is a structural block diagram of the delay duration determination module in the system provided in the embodiments of the present invention;

[0058] Figure 7 This is a structural block diagram of the abnormal sample determination module in the system provided in the embodiments of the present invention;

[0059] Figure 8 This is a structural block diagram of the prediction correction module in the system provided in the embodiment of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0061] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0062] Specifically, a high-efficiency path planning and obstacle avoidance optimization method for robots includes the following steps:

[0063] Step S100: When it is determined that the predicted angle offset of the target robot in the narrow area to be traversed is less than the upper limit of the angle offset, the dynamic influence parameters of its current environment and the passage environment of the narrow area to be traversed are obtained, and the historical cleaning operation database is called.

[0064] The dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighting or calculating the relationship between the two, used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area.

[0065] In this embodiment of the invention, the technical solution is applied to scenarios where underwater robots perform tasks, and is particularly suitable for cleaning operations on underwater structures such as offshore wind farms, seabed platforms, and bridge foundations. This technical solution effectively helps robots perform precise path planning and obstacle avoidance optimization in complex underwater environments, especially when rapidly navigating narrow areas. Besides underwater structure cleaning, this invention is also applicable to underwater exploration, maintenance, and repair robots, and other underwater tasks requiring rapid navigation and obstacle avoidance.

[0066] The target cleaning robot is a tracked or wheeled underwater robot, suitable for cleaning operations between complex underwater structures and in confined spaces. This type of robot has strong mobility, can maintain high work efficiency in relatively complex environments, and can cope with the need for rapid passage through narrow gaps.

[0067] The narrow areas to be traversed specifically refer to the spaces faced by the target cleaning robot when passing through narrow gaps between underwater structures. These narrow areas typically refer to gaps between underwater facilities such as wind turbine foundations or floating platform support structures. The robot must pass through these areas, but due to factors such as water flow and structural interference, passage through these areas is quite difficult.

[0068] Angle offset prediction refers to the predicted angle offset of a target cleaning robot as it traverses a narrow area, based on existing path planning and obstacle avoidance algorithms. This prediction is made in real time by the control system using the robot's current motion state and sensor data (such as IMU, gyroscope, flow sensor, etc.). The angle offset prediction reflects the expected changes in the robot's heading when passing through narrow areas, especially under the influence of environmental factors such as water flow and structural disturbances.

[0069] The upper limit of angular offset refers to the maximum angular offset that the target robot can tolerate when passing through the narrow area. If the actual angular offset exceeds this upper limit, the robot will not be able to safely pass through the area. Both the predicted angular offset and the upper limit of angular offset are based on mature path planning and obstacle avoidance technologies in the existing field. These technologies can accurately predict the robot's trajectory deviation and attitude changes by analyzing the robot's kinematic characteristics and real-time environmental data.

[0070] These predictive methods typically calculate robot behavior in confined spaces using kinematic modeling, real-time data feedback, and environmental perception technologies. These technologies have been widely applied in obstacle avoidance systems across various robotics fields, ensuring robots can perform tasks efficiently and safely in complex environments.

[0071] The dynamic parameters affecting the passage environment include water viscosity and structural vortex disturbance parameters. These two parameters significantly impact the robot's stability when navigating narrow areas. Water viscosity represents the propulsive resistance of the water, and its viscosity significantly affects propulsion efficiency and path stability, especially at low speeds. Structural vortex disturbance parameters describe the vortex effect generated when underwater structures interact with water flow. This vortex disturbance affects the robot's heading and stability, particularly when the robot approaches underwater structures. These two parameters were chosen because they directly influence the robot's motion, and mature measurement and application methods exist in underwater robotics technology.

[0072] Water viscosity parameters and structural vortex disturbance parameters can be obtained using devices such as flow sensors, sonar sensors, and eddy current meters. In existing technologies, water viscosity can be obtained through analysis of water flow velocity, temperature, and water quality, while structural vortex disturbance parameters are typically obtained through fluid dynamics simulations (such as CFD) or field measurements. These methods are well-established and widely used in environmental analysis of underwater robots and underwater structures.

[0073] The historical cleaning operation database records the operational history of all robots performing cleaning tasks in different environments. This database, accumulated through big data technology, provides robots with real-time path planning and obstacle avoidance decision support. The database includes, but is not limited to, the following data types: robot type, propulsion motor performance (e.g., power, usage time), environmental parameters (e.g., water viscosity, eddy interference), task execution status (e.g., success rate, path correction status), and operation logs (e.g., path deviation, obstacle avoidance status). This data will be used to analyze robot performance and provide optimization suggestions for future tasks.

[0074] Furthermore, the robot-specific efficient path planning and obstacle avoidance optimization method also includes the following steps:

[0075] Step S200: Based on the database, select reference cleaning samples of other reference robots that are the same type as the target robot, have the same background parameters of the propulsion motor, and whose historical environment and the dynamic influence parameters of the passage environment corresponding to the narrow area to be crossed in the past are respectively consistent.

[0076] In this embodiment of the invention, "identical target robot type and identical propulsion motor background parameters" specifically means that the target robot and the reference robot are identical in terms of structural type, propulsion system, and working principle. For example, if the target robot is a tracked underwater robot, then the selected reference sample should also be a tracked underwater robot. Furthermore, "identical propulsion motor background parameters" means that the reference robot and the target robot have the same motor power, rated voltage, and service life, as these factors directly affect the robot's propulsion efficiency and control accuracy.

[0077] In this embodiment of the invention, step S200 is specifically implemented by first querying the database to filter out all reference robot data that are identical to the target robot in terms of type and propulsion motor performance parameters. Then, these reference robot data are further filtered to ensure that the environmental conditions in which they perform their tasks are consistent with the dynamic influence parameters of the passage environment of the narrow area to be traversed by the target robot. These environmental parameters include water viscosity, structural eddy interference, etc.

[0078] The reason for choosing such clean reference samples is that the target robot's path planning and obstacle avoidance during task execution requires reference to the performance of other robots that have performed tasks in similar environments. By selecting reference samples that are consistent with the target robot type, propulsion motor performance, and environmental conditions, it can be ensured that these samples have a high degree of similarity to the target robot's performance in path planning and obstacle avoidance. This provides more accurate data support for the target robot's path planning and obstacle avoidance optimization, and avoids path prediction errors caused by selecting irrelevant reference samples.

[0079] Furthermore, the robot-specific efficient path planning and obstacle avoidance optimization method also includes the following steps:

[0080] Step S300: Analyze the reference clean sample and calculate the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from its historical environment.

[0081] Specifically, Figure 2 A flowchart is shown to determine the response hysteresis time corresponding to the reference clean sample.

[0082] The analysis of clean samples, specifically analyzing the response lag time from control signal triggering to the actual generation of thrust by the propulsion motor during the process of entering the narrow area to be traversed from its historical environment, includes the following steps:

[0083] Step S301: Analyze each reference clean sample in sequence, extract the timestamp of the control signal issued by the reference robot control system when instructing the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, and the timestamp of the propulsion motor actually reaching the required thrust for crossing the historical area to be crossed.

[0084] Step S302: Calculate the time difference between the two timestamps as the response lag time of the reference clean sample.

[0085] In this embodiment of the invention, when the robot traverses a narrow area, in order to minimize the time spent in the disturbed area and avoid path deviation or jamming, it typically needs to perform a rapid traversal maneuver. That is, before entering the area, the control system issues a command to drive the propulsion motor to briefly accelerate, quickly generating sufficient thrust to ensure continuous and stable passage. This type of rapid traversal behavior not only requires the robot to have timely response capabilities, but also places higher demands on the dynamic response characteristics of its propulsion system.

[0086] Therefore, during steps S301 and S302, the extracted control signal timestamp corresponds to the moment when the robot control system recognizes the crossing requirement and instructs the propulsion motor to accelerate, while the timestamp of the actual thrust achievement indicates the actual moment when the propulsion motor responds to the acceleration command and outputs the target thrust. The time difference between the two is the response lag time, which reflects the response speed and stability of the propulsion system under temporary acceleration conditions.

[0087] The core research problem of this invention revolves around the response lag time, focusing on identifying whether the propulsion motor performance is sufficient to quickly respond to control commands under disturbed conditions when the robot performs rapid obstacle avoidance maneuvers, and whether environmental disturbances significantly affect this response capability. In other words, this invention focuses on "whether the lag in the propulsion system's response to control signals during a robot's accelerated obstacle avoidance maneuver in a specific dynamic traffic environment may cause deviation," and uses this as a starting point to construct a quantifiable deviation prediction and correction mechanism, thereby improving the reliability and adaptability of the overall obstacle avoidance path.

[0088] During implementation, the control signal timestamp can be obtained from the control system's log data or operation records, and the timestamp when the propulsion motor actually reaches the required thrust can be obtained from the motor's feedback data or detected by the thrust sensing device. The thrust detection device can deduce the thrust achievement time based on the velocity change curves in known torque sensors, pressure sensors, or inertial measurement units (IMUs), and the entire time difference calculation process can be automatically completed by the configured data processing module.

[0089] The response lag time reflects the sensitivity of the propulsion system to control commands. When traversing narrow areas, environmental disturbances to the robot cannot be ignored. If the propulsion motor takes a long time to generate thrust after receiving a control command, the robot may enter a disturbed area before it has acquired sufficient thrust in time, thus affecting its propulsion direction due to lateral disturbances and causing angular deviation. Therefore, this response lag time not only reflects the mechanical performance of the propulsion motor but also indirectly reflects its response reliability under specific dynamic parameters of the passage environment.

[0090] Because traversing narrow areas demands stringent path precision, any lag exceeding the normal control response window can cause angular deviations in the robot's propulsion direction, leading it off the predetermined path. In severe cases, failure to correct these deviations in time can even cause path adjustment to fail. Therefore, calculating and analyzing the response lag time of reference samples quantifies the response time issues that may cause angular deviations under similar conditions. This provides an objective basis for subsequent deviation prediction and correction, ensuring the target robot has sufficient responsiveness to successfully complete the traversal maneuver.

[0091] The timestamps for the control signals, the timestamps for the propulsion motors reaching the required thrust, and the required thrust value itself are all data collected and recorded in the historical cleaning operation database based on past tasks. This database is composed of control command logs, actuator status feedback, and multi-dimensional sensor monitoring results collected in real time by the robot during historical tasks, providing accurate time-series information and action response records. Therefore, in the implementation of this invention, key control events and execution results corresponding to each reference cleaning sample can be directly extracted from the database, ensuring that the extraction of response lag time is based on real historical behavioral data and has sufficient traceability and technical feasibility.

[0092] Furthermore, the robot-specific efficient path planning and obstacle avoidance optimization method also includes the following steps:

[0093] Step S400: Calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and set the reference clean samples with deviation exceeding the preset range as abnormal samples.

[0094] Specifically, Figure 3 A flowchart for identifying anomalous samples is shown.

[0095] The specific steps involved in calculating the deviation of the response lag time of each reference clean sample from the preset standard lag time, and designating reference clean samples with deviations exceeding the preset range as abnormal samples, are as follows:

[0096] Step S401: Obtain the preset standard lag time. The preset standard lag time is the reference response time corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within the preset low threshold.

[0097] Step S402: Calculate the deviation of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time.

[0098] Step S403: The reference clean sample whose deviation exceeds the preset deviation threshold is set as an abnormal sample.

[0099] In this embodiment of the invention, the preset standard lag time, determined through system experiments or large-scale operational data analysis, originates from robot operation data under ideal performance conditions for the propulsion motor and low-disturbance conditions for the dynamic influence parameters of the passage environment. In other words, the preset standard lag time refers to the response time exhibited by the propulsion motor from the triggering of the control command to the actual output of the required thrust when the robot performs a crossing operation under favorable conditions. Under such ideal conditions, when the target robot rapidly crosses from its current environment into a narrow area, its propulsion behavior will not cause additional offset changes beyond the angle offset prediction model constructed by existing technologies. That is, it will not produce angle offsets exceeding the tolerance error of the prediction mechanism itself, thereby ensuring that path following and obstacle avoidance behaviors still have sufficient reliability within the existing model framework.

[0100] This standard lag time is thus recognized as a safety reference benchmark, which can be used to measure whether response delays under non-ideal operating conditions may cause additional angular deviation risks. By comparing the response lag of samples with this benchmark, it is possible to help identify high-risk path deviation trends and provide a reasonable basis for subsequent correction mechanisms.

[0101] The preset standard lag time can be obtained in two ways. The first is to construct a standard experimental environment. In an experimental tank or simulated channel, low-disturbance conditions with controllable water viscosity parameters and structural vortex interference parameters are set. A robot prototype in its factory condition or with well-confirmed performance is selected. Standardized rapid traversal commands are repeatedly executed, and the time difference between the control signal issuance and the actual thrust generated by the propulsion motor is collected. The average value of stable samples is calculated and used as a reference response time. The second method is based on big data sample analysis. Cleaning samples from historical cleaning operation databases are selected where the propulsion motor is within its rated parameter range and the dynamic influence parameters of the passage environment are within a low threshold. The corresponding response lag time is extracted, extreme values ​​are removed, and the average value is calculated as an empirical standard value. The two methods can be selected optimally or used in combination depending on the actual deployment situation to ensure that the standard lag time is verifiable, representative, and statistically robust.

[0102] The deviation magnitude in step 402 is used to quantify the relative difference between the response lag time of each reference clean sample during actual crossing operations and the reference value under ideal conditions. This deviation magnitude reflects the degree of change in the robot's controlled response performance during acceleration in a confined area, essentially revealing whether the stability of the robot's propulsion action is affected by the state of its propulsion motor or environmental disturbances. The larger the deviation magnitude, the weaker the robot's actual response capability under these conditions, potentially increasing the risk of insufficient angle offset prediction or crossing failure. Therefore, screening based on this deviation magnitude can effectively identify abnormal samples with unstable or delayed execution capabilities, providing a data basis and judgment for subsequent correction of angle offset predictions.

[0103] Furthermore, the robot-specific efficient path planning and obstacle avoidance optimization method also includes the following steps:

[0104] Step S500: Determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, calculate the adjustment factor based on the deviation magnitude of all abnormal samples, apply the adjustment factor to the angle offset prediction, obtain the corrected angle offset prediction, and determine whether the target robot is allowed to pass through the narrow area to be passed.

[0105] Specifically, Figure 4 A flowchart is shown for correcting the initial angular offset prediction.

[0106] The process of determining whether the proportion of abnormal samples in all clean reference samples exceeds a preset ratio includes the following steps: If so, an adjustment factor is calculated based on the deviation magnitude of all abnormal samples, and this adjustment factor is applied to the angle offset prediction to obtain the corrected angle offset prediction. Based on this, it is determined whether the target robot is allowed to pass through the narrow area to be traversed.

[0107] Step S501: Calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and determine whether it exceeds the preset ratio threshold.

[0108] Step S502: If the deviation exceeds the limit, obtain the deviation of the response lag time of all abnormal samples relative to the preset standard lag time, and set the average value of all deviations as the adjustment factor.

[0109] Step S503: Call the preset angle offset prediction correction function to correct the initial angle offset prediction based on the adjustment factor, and obtain the corrected angle offset prediction.

[0110] Step S504: Compare the corrected angle offset prediction with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed. If it is not less than the upper limit of the angle offset, it is determined that it cannot pass through.

[0111] The angle offset prediction correction function is:

[0112] ;

[0113] in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time;

[0114] It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

[0115] In this embodiment of the invention, the significance of setting step S501 is to identify whether the environment in which the current target robot may be located has significant non-ideal characteristics by quantifying the proportion of abnormal samples among all reference clean samples. If the proportion of abnormal samples exceeds a preset ratio threshold, it indicates that under similar propulsion motor performance and dynamic influence parameters of the passage environment, a large proportion of robots exhibit significantly larger response lags when performing rapid crossing actions, which is highly likely to cause unexpected angle deviations. This preset ratio threshold can be obtained from system experiments or historical large-scale operation data analysis, and is usually taken within the empirical upper limit range that has a statistically significant impact on the angle deviation prediction results, ensuring that the judgment logic is both sensitive and generalizable.

[0116] Using the average deviation of all abnormal samples as an adjustment factor, the systemic delay characteristics of robot propulsion behavior under non-ideal conditions can be effectively captured while maintaining overall stability. The deviation essentially reflects the degree of efficiency reduction of the propulsion motor in actual working conditions compared to the ideal state. This efficiency reduction can easily lead to lag in the execution of angle control commands during the robot's rapid traversal of narrow areas, resulting in problems such as steering delay and amplified angle deviation along the physical path.

[0117] When traversing narrow areas, robots need to accelerate rapidly and control their posture precisely. If the propulsion motors have a significant response delay, their response to angle adjustment commands will also lag, causing deviations from the target angle on the actual travel path. Since existing angle deviation prediction models are often built based on ideal response assumptions and fail to dynamically reflect the time delay characteristics of the propulsion motor coupled with environmental interference factors, there is a risk of prediction distortion.

[0118] By converting the average magnitude of the response lag deviation into an adjustment factor and embedding it into the angle deviation prediction function, this mechanism essentially uses this average deviation to measure the overall weakening of propulsion behavior control under the current operating conditions. This, in turn, proportionally corrects the predicted angle deviation, making it closer to the actual possible attitude deviation. This mechanism not only compensates for the insufficient adaptability of existing prediction models under non-ideal conditions but also dynamically adapts to changes in the intensity of environmental disturbances and propulsion performance, forming a flexible adjustment structure. This better addresses the path deviation risk caused by response lag during rapid crossing operations, effectively supplementing and enhancing traditional static model-based prediction methods.

[0119] Step S504 is a crucial determination of the correction result. The system compares the corrected angle offset prediction with the set upper limit of angle offset. If the prediction is within an acceptable range, the robot is considered to be able to safely complete the crossing action under the current state, avoiding misjudgment as impassable due to excessive conservatism. If the prediction exceeds the threshold, it indicates a high risk of offset under the current working condition, thus promptly preventing the crossing action from being executed. In particular, this step not only evaluates the path-following capability but also has a direct impact on maintaining obstacle avoidance accuracy: in narrow areas, the robot is more sensitive to angle control when executing obstacle avoidance strategies. If the angle offset exceeds the threshold, it is highly likely to cause obstacle avoidance failure or collision risk. By setting this step, the path planning system can achieve perceptual response adjustment to minor performance anomalies and environmental disturbances, improving its judgment ability under boundary conditions and ensuring the safety and robustness of path following, obstacle avoidance behavior, and overall task execution.

[0120] Furthermore, Figure 5 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0121] In another preferred embodiment of the present invention, a high-efficiency path planning and obstacle avoidance optimization system for robots includes:

[0122] The data acquisition module 100 is used to acquire the current environment of the target robot and the dynamic influence parameters of the passage environment of the narrow area to be traversed when it is determined that the predicted angular offset of the target robot in the narrow area to be traversed is less than the upper limit of the angular offset, and to call the historical cleaning operation database.

[0123] Furthermore, the robot-specific high-efficiency path planning and obstacle avoidance optimization system also includes:

[0124] The reference sample acquisition module 200 is used to filter reference cleaning samples of other reference robots that are consistent with the target robot type, have the same propulsion motor background parameters, and whose historical environment and the corresponding dynamic influence parameters of the passage environment of the narrow area to be traversed are consistent, based on the database. The dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighted processing or functional relationship calculation of the two, which are used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area.

[0125] Furthermore, the robot-specific high-efficiency path planning and obstacle avoidance optimization system also includes:

[0126] The lag time determination module 300 is used to analyze the reference clean sample and count the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from the historical environment.

[0127] Specifically, Figure 6 The diagram shows the structural block diagram of the delay duration determination module 300 in the system provided by the embodiment of the present invention.

[0128] In a preferred embodiment provided by the present invention, the delay duration determination module 300 specifically includes:

[0129] The timestamp acquisition unit 301 is used to sequentially parse each reference clean sample, extract the timestamp of the control signal issued by the reference robot control system when instructing the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, and the timestamp of the propulsion motor actually reaching the required thrust to cross the historical area to be crossed.

[0130] The time difference calculation unit 302 is used to calculate the time difference between two timestamps, which serves as the response lag time of the reference clean sample.

[0131] Furthermore, the robot-specific high-efficiency path planning and obstacle avoidance optimization system also includes:

[0132] The abnormal sample determination module 400 is used to calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and to set the reference clean sample whose deviation exceeds the preset range as an abnormal sample.

[0133] Specifically, Figure 7 The diagram shows the structural block diagram of the abnormal sample determination module 400 in the system provided by an embodiment of the present invention.

[0134] In a preferred embodiment of the present invention, the abnormal sample determination module 400 specifically includes:

[0135] The standard duration acquisition unit 401 is used to acquire a preset standard lag duration, which is the reference response duration corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within a preset low threshold.

[0136] The deviation magnitude calculation unit 402 is used to calculate the deviation magnitude of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation magnitude is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time.

[0137] The abnormal sample setting unit 403 is used to set reference clean samples whose deviation exceeds a preset deviation threshold as abnormal samples.

[0138] Furthermore, the robot-specific high-efficiency path planning and obstacle avoidance optimization system also includes:

[0139] The prediction correction module 500 is used to determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, it calculates an adjustment factor based on the deviation magnitude of all abnormal samples, applies the adjustment factor to the angle offset prediction, obtains the corrected angle offset prediction, and determines whether the target robot is allowed to pass through the narrow area to be passed.

[0140] Specifically, Figure 8 A structural block diagram of the prediction correction module 500 in the system provided by an embodiment of the present invention is shown.

[0141] In a preferred embodiment provided by the present invention, the prediction correction module 500 specifically includes:

[0142] The ratio judgment unit 501 is used to calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and to determine whether it exceeds the preset ratio threshold.

[0143] The adjustment factor determination unit 502 is used to obtain the deviation of the response lag time of all abnormal samples from the preset standard lag time if the deviation exceeds the limit, and set the average value of all deviations as the adjustment factor.

[0144] The prediction correction unit 503 is used to call a preset angle offset prediction correction function to correct the initial angle offset prediction based on the adjustment factor, so as to obtain the corrected angle offset prediction.

[0145] The corrected prediction unit is used to compare the corrected angle offset prediction with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed; if it is not less than the upper limit, it is determined that it cannot pass through.

[0146] The angle offset prediction correction function is:

[0147] ;

[0148] in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time;

[0149] It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

[0150] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0151] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0154] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A highly efficient path planning and obstacle avoidance optimization method specifically for robots, characterized in that, The method includes: When it is determined that the predicted angular offset of the target robot in the narrow area to be traversed is less than the upper limit of the angular offset, the dynamic influence parameters of its current environment and the passage environment of the narrow area to be traversed are obtained, and the historical cleaning operation database is called. Based on the database, reference cleaning samples of other reference robots that are the same type as the target robot, have the same background parameters of the propulsion motor, and whose historical environment and the dynamic influence parameters of the passage environment corresponding to the narrow area to be crossed in the past are respectively consistent. By analyzing clean samples, we statistically analyzed the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from the historical environment. Calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and set the reference clean samples with deviation exceeding the preset range as abnormal samples; Determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, calculate an adjustment factor based on the deviation magnitude of all abnormal samples, apply the adjustment factor to the angle offset prediction, obtain the corrected angle offset prediction, and determine whether the target robot is allowed to pass through the narrow area to be traversed.

2. The efficient path planning and obstacle avoidance optimization method for robots according to claim 1, characterized in that, The dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighting or calculating the relationship between the two, used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area.

3. The efficient path planning and obstacle avoidance optimization method for robots according to claim 2, characterized in that, Analyzing the clean sample, the steps involved in the process of moving from its historical environment into the historical narrow passage to be traversed, from the triggering of the control signal to the actual generation of the required thrust by the propulsion motor, include: Each reference clean sample is analyzed sequentially to extract the timestamp of the control signal issued by the reference robot control system when it instructs the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, as well as the timestamp of the propulsion motor actually reaching the required thrust to cross the historical narrow area to be crossed. Calculate the time difference between the two timestamps as the response lag time for the reference clean sample.

4. The efficient path planning and obstacle avoidance optimization method for robots according to claim 3, characterized in that, The steps of calculating the deviation of the response lag time of each reference clean sample from the preset standard lag time, and setting the reference clean samples with deviations exceeding the preset range as abnormal samples, include: Obtain a preset standard lag time, which is the reference response time corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within a preset low threshold. Calculate the deviation of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time. Reference clean samples whose deviation exceeds a preset deviation threshold are set as abnormal samples.

5. The efficient path planning and obstacle avoidance optimization method for robots according to claim 4, characterized in that, The steps for determining whether the proportion of abnormal samples in all clean reference samples exceeds a preset ratio, and if so, to calculate an adjustment factor based on the deviation magnitude of all abnormal samples, apply the adjustment factor to the angle offset prediction, obtain the corrected angle offset prediction, and determine whether the target robot is allowed to pass through the narrow area to be traversed include: Calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and determine whether it exceeds a preset ratio threshold; If the deviation exceeds the limit, obtain the deviation of the response lag time of all abnormal samples relative to the preset standard lag time, and set the average value of all deviations as the adjustment factor; The preset angle offset prediction correction function is called to correct the initial angle offset prediction based on the adjustment factor, and the corrected angle offset prediction is obtained. The corrected angle offset prediction is compared with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed. If it is not less than the upper limit of the angle offset, it is determined that it cannot pass through.

6. The efficient path planning and obstacle avoidance optimization method for robots according to claim 5, characterized in that, The angle offset prediction correction function is: ； in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time; It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

7. A high-efficiency path planning and obstacle avoidance optimization system specifically for robots, characterized in that, The system includes: a data acquisition module, a reference sample acquisition module, a lag duration determination module, an outlier sample determination module, and a prediction correction module, wherein: The data acquisition module is used to acquire the current environment of the target robot and the dynamic influence parameters of the passage environment of the narrow area to be traversed when it is determined that the predicted angular offset of the target robot in the narrow area to be traversed is less than the upper limit of the angular offset, and to call the historical cleaning operation database. The reference sample acquisition module is used to filter out reference cleaning samples of other reference robots that are consistent with the target robot type, have the same propulsion motor background parameters, and whose historical environment and the corresponding dynamic influence parameters of the passage environment of the narrow area to be traversed are consistent, based on the database. The dynamic influence parameters of the passage environment include a weighted combination of water viscosity parameters and structural vortex disturbance parameters, or other combinations obtained by weighting the two or by calculating the functional relationship, which are used to characterize the fluid resistance and disturbance coupling effect experienced by the robot in a specific area. The lag time determination module is used to analyze the reference clean sample and calculate the response lag time from the control signal trigger to the actual thrust generated by the propulsion motor during the process of entering the narrow area to be crossed from its historical environment. The abnormal sample determination module is used to calculate the deviation of the response lag time of each reference clean sample from the preset standard lag time, and to set the reference clean sample whose deviation exceeds the preset range as an abnormal sample. The prediction correction module is used to determine whether the proportion of abnormal samples in all reference clean samples exceeds a preset ratio. If so, it calculates an adjustment factor based on the deviation magnitude of all abnormal samples, applies the adjustment factor to the angle offset prediction, obtains the corrected angle offset prediction, and determines whether the target robot is allowed to pass through the narrow area to be traversed.

8. The high-efficiency path planning and obstacle avoidance optimization system for robots according to claim 7, characterized in that, The delay duration determination module specifically includes: The timestamp acquisition unit is used to sequentially parse each reference clean sample, extract the timestamp of the control signal issued by the reference robot control system when instructing the propulsion motor to perform the action of quickly crossing from the historical environment into the historical narrow area to be crossed, and the timestamp of the propulsion motor actually reaching the required thrust to cross the historical narrow area to be crossed. The time difference calculation unit is used to calculate the time difference between two timestamps, which serves as the response lag time of the reference clean sample.

9. The high-efficiency path planning and obstacle avoidance optimization system for robots according to claim 8, characterized in that, The abnormal sample determination module specifically includes: The standard duration acquisition unit is used to acquire a preset standard lag duration, which is the reference response duration corresponding to the robot control system triggering the propulsion motor to generate the required thrust under the condition that the propulsion motor is in rated performance state and the dynamic influence parameters of the traffic environment are within a preset low threshold. The deviation magnitude calculation unit is used to calculate the deviation magnitude of the response lag time of each reference clean sample relative to the preset standard lag time. The deviation magnitude is the ratio of the difference between the response lag time of the reference clean sample and the preset standard lag time to the preset standard lag time. The abnormal sample setting unit is used to set reference clean samples whose deviation exceeds a preset deviation threshold as abnormal samples.

10. The high-efficiency path planning and obstacle avoidance optimization system for robots according to claim 9, characterized in that, The prediction correction module specifically includes: The ratio judgment unit is used to calculate the ratio of the number of abnormal samples to the total number of reference clean samples, and to determine whether it exceeds the preset ratio threshold. The adjustment factor determination unit is used to obtain the deviation of the response lag time of all abnormal samples from the preset standard lag time if the deviation exceeds the limit, and set the average value of all deviations as the adjustment factor. The prediction correction unit is used to call the preset angle offset prediction correction function to correct the initial angle offset prediction based on the adjustment factor, so as to obtain the corrected angle offset prediction. The corrected prediction unit is used to compare the corrected angle offset prediction with the upper limit of the angle offset. If it is less than the upper limit of the angle offset, it is determined that the target robot can pass through the narrow area to be passed; if it is not less than the upper limit, it is determined that it cannot pass through. The angle offset prediction correction function is: ； in, This refers to the corrected angular offset prediction. This refers to the uncorrected angular offset prediction. This refers to the total number of abnormal samples. It refers to the first The response lag time corresponding to each abnormal sample This refers to the preset standard lag time; It refers to the first The deviation of the response lag time for each abnormal sample from the preset standard lag time. Indicates the first An outlier sample is only included in the relative deviation ratio if its response lag time exceeds the preset standard lag time. If it does not exceed the preset standard lag time, it is considered to have zero contribution. This is used to exclude cases where the response lag time is earlier than or exactly equal to the preset standard lag time. This refers to the adjustment factor. This refers to the weighting factor of the adjustment factor, and Greater than 0.

Images