A method for maintaining ad hoc network links during robot movement

Abstract

The application discloses a kind of self-organizing network link maintenance method in the movement of robot, specifically relates to ultra-dense networking technical field, including: the physical layer parameter of communication link and the motion state information of local and adjacent robot are collected;Signal quality sub-factor is constructed based on physical layer parameter, motion trend sub-factor is constructed based on relative distance and angle change rate, and integrated link quality factor is obtained by dynamic weighting fusion;According to position coordinates and velocity vector, link residual lifetime is predicted;Residual lifetime is compared with the multiple threshold value of preestablished, and hierarchical link maintenance mechanism is executed: when being in mild risk, power control is carried out, when being in moderate risk, backup path is established but not switched, when being in emergency risk, immediately switch to backup path and start data buffer.The application improves the link stability and resource utilization efficiency of robot self-organizing network in high dynamic environment by multidimensional link evaluation and hierarchical response strategy.

Description

Technical Field

[0001] This invention relates to the field of ultra-dense networking technology, and more specifically, to a method for maintaining self-organizing network links during robot movement. Background Technology

[0002] This invention relates to the field of ultra-dense networking technology, and in particular to link maintenance technology for robot mobile ad hoc networks. As a type of decentralized distributed communication network, robot ad hoc networks are widely used in scenarios such as multi-robot collaborative operations, industrial inspection, emergency rescue, and outdoor exploration. The core requirement is to maintain the continuity and stability of the communication link during the dynamic movement of a group of robots, and to ensure the reliability of data transmission.

[0003] Existing technologies attempt to introduce link quality prediction, but most only consider physical layer communication parameters (such as signal-to-noise ratio and bit error rate), without incorporating the relative motion characteristics during robot movement. However, when robots operate in swarms, the speed, heading, and relative distance of each node are the core factors causing dynamic changes in the link topology. Single signal quality prediction is insufficient to accurately determine the long-term stability of the link. Furthermore, existing link maintenance mechanisms lack tiered risk handling strategies. Faced with different degrees of link deterioration, they all adopt the same response, either excessively consuming communication resources or lacking effective contingency plans in case of emergency link deterioration, resulting in low network resource utilization and long link interruption recovery times.

[0004] Existing methods for maintaining links in robot ad hoc networks can meet practical application requirements, but they still have some drawbacks, such as a single dimension for link quality assessment: most methods rely only on physical layer parameters (such as RSSI) or a certain type of data in motion information, failing to comprehensively consider the combined effects of signal quality and motion trends, resulting in inaccurate link status judgment and easy misjudgment or omission. Lack of dynamic weight adjustment mechanism: When the robot's motion state changes drastically (such as acceleration, turning, high-speed movement), the factors affecting link quality will change. Existing technology has failed to dynamically adjust the weights of signal quality and motion trend according to the intensity of motion, resulting in poor adaptability of the evaluation model. Link hold-up strategies suffer from delayed response: Existing link hold-up mechanisms are mostly triggered by a single threshold, lacking the ability to handle different levels of risk. When link quality deteriorates, they often directly trigger route switching or power adjustments, failing to take differentiated measures based on the degree of risk, resulting in wasted resources or untimely response. Insufficient accuracy in predicting remaining link lifetime: Existing prediction models are mostly based on simple distance change rates, without taking into account factors such as relative angle changes and motion trends. The prediction results have large errors and are difficult to meet the real-time requirements in highly dynamic environments.

[0005] Existing technologies still have significant shortcomings in comprehensive link quality assessment, dynamic weight adjustment, hierarchical response mechanisms, lifetime prediction accuracy, and data cache recovery, making it difficult to meet the communication requirements of robot ad hoc networks with high dynamics and high reliability. Therefore, there is an urgent need for a link maintenance method that can integrate multi-source information, dynamically adjust assessment weights, hierarchically respond to link risks, and has data caching capabilities to improve communication stability and task execution efficiency during robot movement. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, the present invention provides a method for maintaining ad hoc network links during robot movement, which solves the problems mentioned in the background art through the following scheme.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for maintaining ad-hoc network links during robot movement, comprising: S1: Collect the physical layer parameters of the current communication link, and simultaneously collect the motion state information of the local robot and nearby robots; the physical layer parameters include received signal strength indication, signal-to-noise ratio, and bit error rate; the motion state information includes velocity vector, acceleration, heading angle, and position coordinates; S2: Construct a signal quality sub-factor based on the physical layer parameters, construct a motion trend sub-factor based on the relative distance change rate and relative angle change rate in the motion state information, and perform weighted fusion of the signal quality sub-factor and the motion trend sub-factor to obtain a comprehensive link quality factor. The weighted fusion weight is dynamically adjusted according to the robot's current motion state. S3: Calculate the relative motion speed based on the position coordinates and velocity vectors of the local robot and neighboring robots, and predict the remaining link lifetime by combining the maximum communication distance of wireless communication; S4: Compare the remaining link lifetime with a preset first threshold, a second threshold, and an emergency threshold. If the emergency threshold is less than the first threshold, and the first threshold is less than the second threshold, then execute a tiered link maintenance mechanism. S41: When the remaining lifetime of the link is greater than or equal to the second threshold, maintain the current link state; S42: When the remaining lifetime of the link is between the first threshold and the second threshold, it is determined to be a mild risk. A power control command is sent to the peer robot to perform power control in order to maintain the link. S43: When the remaining lifetime of the link is between the emergency threshold and the first threshold, it is determined to be of medium risk, triggering route pre-switching, establishing a backup path but not immediately switching the primary path; S44: When the remaining lifetime of the link is lower than the emergency threshold and the overall link quality factor decreases by more than a preset percentage per unit time, it is determined to be an emergency risk, and the system immediately switches to the backup path and starts data caching.

[0008] The technical effects and advantages of this invention are as follows: 1. This invention constructs a signal quality sub-factor by integrating physical layer parameters (RSSI, SNR, BER) and constructs a motion trend sub-factor by combining the relative distance change rate and relative angle change rate in motion state information. A weighted fusion method is then used to obtain the comprehensive link quality factor. Compared to single-dimensional evaluation methods, this approach can more comprehensively and accurately reflect the link status, improving the robustness and reliability of link quality assessment. 2. This invention dynamically adjusts the fusion weights of the signal quality sub-factor and the motion trend sub-factor based on the intensity of the robot's current motion (speed and acceleration). In high-dynamic motion scenarios, the weight of the motion trend increases; in low-speed or stationary states, the signal quality weight rebounds, improving the adaptability of link determination in complex motion environments. 3. This invention divides the remaining link lifetime into multiple risk levels (mild, moderate, and critical) and adopts differentiated processing strategies such as power control, route pre-switching, emergency switching, and data caching accordingly. In the case of mild risk, only power is adjusted to avoid resource waste caused by frequent switching. In the case of moderate risk, backup paths are established in advance to shorten the interruption response time. In the case of critical risk, switching is performed immediately and data caching is initiated. This hierarchical mechanism improves the efficient utilization of communication resources and link interruption recovery. 4. Based on the position coordinates, velocity vectors, and relative radial velocities of the local robot and neighboring robots, and combined with the maximum wireless communication distance, this invention uses a linear motion assumption model to predict the remaining link lifetime. Compared with the traditional method that only relies on the distance change rate, it introduces relative velocity vectors and motion trend information, resulting in more accurate predictions. It is especially suitable for real-time link assessment and decision-making in highly dynamic mobile environments. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the overall method structure of the present invention.

[0010] Figure 2 This is a schematic diagram of data acquisition in S1 of the present invention.

[0011] Figure 3 This is a schematic diagram of the construction of the comprehensive link quality factor in S2 of the present invention.

[0012] Figure 4 This is a schematic diagram of the link remaining lifetime prediction in S3 of the present invention.

[0013] Figure 5This is a schematic diagram of the hierarchical link preservation mechanism of S4 in this invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figures 1-5 As shown, this embodiment of the invention provides a method for maintaining ad-hoc network links during robot movement, the method comprising: S1: Collect the physical layer parameters of the current communication link, and simultaneously collect the motion state information of the local robot and nearby robots; the physical layer parameters include received signal strength indication, signal-to-noise ratio, and bit error rate; the motion state information includes velocity vector, acceleration, heading angle, and position coordinates; S2: Construct a signal quality sub-factor based on the physical layer parameters, construct a motion trend sub-factor based on the relative distance change rate and relative angle change rate in the motion state information, and perform weighted fusion of the signal quality sub-factor and the motion trend sub-factor to obtain a comprehensive link quality factor. The weighted fusion weight is dynamically adjusted according to the robot's current motion state. S3: Calculate the relative motion speed based on the position coordinates and velocity vectors of the local robot and neighboring robots, and predict the remaining link lifetime by combining the maximum communication distance of wireless communication; S4: Compare the remaining link lifetime with a preset first threshold, a second threshold, and an emergency threshold. If the emergency threshold is less than the first threshold, and the first threshold is less than the second threshold, then execute a tiered link maintenance mechanism. S41: When the remaining lifetime of the link is greater than or equal to the second threshold, maintain the current link state; S42: When the remaining lifetime of the link is between the first threshold and the second threshold, it is determined to be a mild risk. A power control command is sent to the peer robot to perform power control in order to maintain the link. S43: When the remaining lifetime of the link is between the emergency threshold and the first threshold, it is determined to be of medium risk, triggering route pre-switching, establishing a backup path but not immediately switching the primary path; S44: When the remaining lifetime of the link is lower than the emergency threshold and the overall link quality factor decreases by more than a preset percentage per unit time, it is determined to be an emergency risk, and the system immediately switches to the backup path and starts data caching.

[0016] In S1, the physical layer parameters of the current communication link are collected, and motion state information of the local robot and nearby robots is also collected. S11: Triggering data acquisition and synchronization Data acquisition tasks are triggered periodically with a preset period T, or real-time acquisition is triggered when the communication module detects packet loss or retransmission events in the link; to ensure the timing consistency of subsequent fusion computing, the data collected by the robot operating system ROS is labeled with a unified time tag through the distributed clock synchronization protocol of the robot. Further explanation is needed regarding the local and neighboring robots. The local robot is controlled by the Explorer 5 remote controller, which is a waterproof industrial-grade remote controller based on an embedded operating system. It supports customer-defined communication and control protocols and uses a self-organizing network communication method. It features an aluminum alloy shell for excellent heat dissipation, a silicone-wrapped grip, and a shoulder strap for easy operation. It also supports bidirectional transmission: images and information captured by the front-end are transmitted to the back-end robot remote controller via the front-end transmitter. The back-end robot remote controller can then send control commands to the robot, enabling bidirectional data transmission within the robot's intelligent communication system. High-definition, high-brightness screen: It adopts a 10.1-inch high-definition, high-brightness screen that is visible even in sunlight. The latency of simulated video images is no more than 0.3 seconds, and the latency of the network camera is no more than 0.4 seconds. The data fed back from the front end is visible in real time. The protection level adopts IP67 protection design, which is convenient for use in harsh environments.

[0017] S12: Collection of physical layer parameters of current communication link This machine reads real-time physical layer data from the communication links of neighboring robots through the wireless communication modules of the Explorer 5 remote controller, such as the Wi-Fi module and the driver interface for self-organizing network communication. Received Signal Strength Indicator (RSSI) Acquisition: By reading the network interface card (NIC) register, the average power value of the received data frames within a unit period T is obtained; Signal-to-noise ratio (SNR) acquisition: The power ratio of signal to noise is read from the underlying status register of the communication baseband chip. It is usually estimated by evaluating the channel quality of the preamble or pilot signal. Bit Error Rate (BER) Acquisition: Calculated by statistically analyzing the proportion of erroneous bits in the physical layer decoding to the total number of bits; by parsing the check sequence in the physical layer header of the data frame, accumulating the number of erroneous frames over a period of time, and calculating the instantaneous BER.

[0018] S13: Acquisition of machine motion status information This machine collects motion status information data through its internal sensor bus and robot operating system nodes: Velocity vector and acceleration acquisition: Read the inertial measurement unit (IMU) data, fuse it using Kalman filtering, and obtain the three-axis acceleration of the machine in the carrier coordinate system. Simultaneously, the solution results from the wheel encoder are read to obtain the velocity vector of the machine in the global coordinate system. ; Heading angle Data Acquisition: The current heading angle of the machine is obtained by reading the attitude calculation results after fusing electronic compass and IMU data. It is usually defined as the angle between the north direction and the X-axis of the global coordinate system; Position coordinate acquisition: Latitude and longitude coordinates are obtained through GNSS satellite navigation and converted into the robot's local position coordinates. .

[0019] S14: Acquisition of motion state information of neighboring robots The local unit sends status request commands to neighboring robots via a wireless communication link, either through broadcasting or polling. The motion controller of the neighboring robot responds to the request, packages its own motion status information—velocity vector, acceleration, heading angle, and position coordinates—into a standard format data frame, and transmits it back to the local unit via the communication link. Upon receiving the data, the local unit distinguishes different neighboring robots based on the MAC address in the data frame and stores it in its local dynamic neighbor information table.

[0020] S15: Data Filtering and Alignment Due to sensor noise and communication delay, the raw data collected may contain errors. Sliding window averaging filtering or first-order low-pass filtering algorithms are used to process physical parameters such as RSSI, SNR, and BER. At the same time, linear interpolation algorithms are used to correct motion state data with mismatched timestamps of neighboring robots caused by network delay, ensuring that the data calculated in subsequent S2 calculations have the same time domain reference.

[0021] In S2, the process of constructing signal quality sub-factors and motion trend sub-factors based on physical layer parameters and motion state information, and obtaining a comprehensive link quality factor through dynamic weighted fusion, specifically includes the following sub-steps: S21: Signal quality sub-factor Construction The signal quality sub-factor is used to quantify the instantaneous quality of the current communication link. It is calculated by combining the physical layer parameters RSSI, SNR, and BER. Considering the physical meaning and dimensional differences of each parameter, it is first normalized and then obtained by weighted summation. RSSI Normalization: Sets the effective operating range of RSSI For ranges from -100dBm to -30dBm, a linear mapping function is used:

[0022] And truncate the results to interval; SNR normalization: sets the effective range of SNR. For example, from 0dB to 30dB, a similar linear mapping is used:

[0023] Similarly, truncate the result to... interval; BER normalization: A lower bit error rate is generally better, and the relationship is non-linear. Negative exponential or logarithmic mapping can be used; this embodiment uses a negative exponential function.

[0024] in For adjustment coefficients (e.g.) This means that when the BER is small, the factor is close to 1, and when the BER increases, the factor decreases rapidly. Weighted fusion: Combining three normalization factors according to preset weights yields signal quality sub-factors.

[0025] in The weights can be pre-calibrated according to the actual channel characteristics, such as... .

[0026] S22: Calculation of relative motion parameters Based on the motion state information (position coordinates, velocity vector, heading angle) of the local robot and its neighboring robots, calculate the relative distance change rate and the relative angle change rate; Relative distance change rate First, calculate the Euclidean distance between two adjacent robot coordinate nodes. Then, the relative radial velocity is calculated using the velocity vector:

[0027] in and These represent the velocity vectors of the local robot and the neighboring robot, respectively. Time indicates that the two nodes are far apart. It indicates that they are close to each other.

[0028] Relative angle change rate Define the machine's heading angle. for Direction angle of the line connecting the two nodes The absolute value of the difference, i.e. ,in Relative angle change rate We obtain the following through numerical differentiation:

[0029] in The sampling period is This reflects how quickly the machine's heading changes relative to the direction of the node connection. A larger value indicates a more drastic change in the direction of movement, which may lead to link instability.

[0030] S23: Motion Trend Sub-Factor Construction The motion trend sub-factor is used to assess the link deterioration trend caused by relative node motion, and is determined by the relative distance change rate. and relative angle change rate A joint decision.

[0031] Distance change rate mapping: mapping relative distance change rate Mapped to stability components ,use The function will Mapping to the [0,1] interval:

[0032] in As the slope control parameter, it makes The larger Approaching 0, The smaller Approaching 1; when hour, ; Angular change rate mapping: mapping relative angular change rate Mapped to stability components The same method is used. The function will Mapped to Interval:

[0033] in The slope is a control parameter that maximizes the absolute value of the rate of change of the angle. The smaller the value, the less stable the response; Comprehensive motion trend sub-factors ,in The balancing coefficient can be preset according to the application scenario, for example... It places more emphasis on changes in distance.

[0034] S24: Dynamic Weight Calculation Weighted fusion weight Based on the robot's current motion state, i.e., speed magnitude and the magnitude of acceleration Dynamic adjustments are made to highlight key influencing factors under different motion patterns; Define the intensity index of exercise :

[0035] in and The preset maximum speed and maximum acceleration represent the robot's ultimate capabilities. For example, weighting coefficients , , ; Determine the signal quality factor weight When the robot is moving vigorously, the movement trend has a greater impact on future links, so the weight of the signal quality factor should be reduced; conversely, when stationary or moving at low speeds, signal quality plays a dominant role, and linear adjustment should be used.

[0036] in As the benchmark weight, To adjust the step size and limit ; Determine motion trend factors weight for: .

[0037] S25: Weighted fusion yields the comprehensive link quality factor

[0038] Signal quality sub-factor and motion trend sub-factor Fusion by dynamic weights:

[0039] in The range of values ​​is A higher value indicates better and more stable link quality. This factor will serve as an important reference for subsequent prediction of remaining link lifetime and hierarchical maintenance mechanisms.

[0040] In S3, the relative motion speed is calculated based on the position coordinates and velocity vectors of the local robot and the neighboring robot, and the remaining link lifetime is predicted by combining the maximum communication distance of wireless communication. Based on acquiring the motion state information S1 of the local robot and neighboring robots and completing the construction of the comprehensive link quality factor S2, the remaining lifetime of the current communication link is predicted based on the relative kinematics model. The remaining lifetime of the link is defined as the time elapsed from the current moment until the distance between the two robots exceeds the maximum effective distance of wireless communication due to relative motion. Specifically, it includes the following sub-steps: S31: Extract motion state parameters (1) Extract the current time t of the local machine from the data collected from S1 and stored in the dynamic neighbor information table: Position coordinates ; velocity vector ; (2) and the target's neighboring robots: Position coordinates ; velocity vector ; To ensure prediction accuracy, data aligned using the S15 filter is required.

[0041] S32: Calculate relative motion parameters (1) Relative position vector: ; (2) Current Euclidean distance : ; (3) Relative velocity vector: ; (4) Relative radial velocity, i.e., the velocity component along the line connecting the two sides:

[0042] in Time indicates that the two nodes are far apart. Time indicates that they are close to each other. This indicates that the relative radial velocities of the two nodes remain constant.

[0043] S33: Obtain the maximum communication distance Read the system's preset maximum effective wireless communication range This parameter can be preset according to the theoretical specifications of the communication equipment, such as by calibration based on actual field test results; It can be adaptively adjusted according to the current environment, such as dynamically correcting based on historical signal quality data. This embodiment uses static preset values ​​to ensure real-time performance.

[0044] S34: Predict the remaining link lifetime

[0045] Based on current distance and relative radial velocity The linear motion assumption is adopted, that is, it is assumed that the two nodes maintain their current velocity vectors unchanged during the prediction time, and the remaining link lifetime is calculated. Scenario 1: Relatively far away or stationary ( ) If the two nodes are far apart or their relative radial velocity is zero, the link will break at some point in the future. The remaining lifetime of the link is calculated by the following formula:

[0046] Among them, when When the link is interrupted, the remaining time to survival is considered to be 0. This is typically used for prediction of established links. ; like If the distance remains constant, the link can be maintained indefinitely, theoretically. In actual processing, a maximum value can be assigned, such as 106 seconds, used for subsequent threshold comparisons.

[0047] Scenario 2: Relatively close ( ) As two nodes approach each other, the distance gradually decreases, communication quality improves, and the link will not be interrupted due to excessive distance. However, considering the possibility of changing direction during actual movement, for uniform handling, the remaining link lifetime can be set to a relatively large constant, for example... This indicates that there is currently no risk of interruption.

[0048] Directional change prediction can be introduced: If the subsequent movement trend may turn away, it needs to be corrected by combining the movement trend sub-factor. The remaining link lifetime is calculated by the following formula: ,For example 104 seconds.

[0049] S35: Output and Storage The calculated The corresponding neighboring robots are identified and associated, and stored as input parameters for the subsequent hierarchical link maintenance mechanism S4; simultaneously, the calculated value of the current time t is recorded for monitoring in the next cycle. The changing trend.

[0050] In S4, the remaining link lifetime is compared with a preset first threshold, a second threshold, and an emergency threshold. The emergency threshold is less than the first threshold, the first threshold is less than the second threshold, and a tiered link maintenance mechanism is executed. Based on the prediction of remaining lifetime of the S3 link and the calculation of the comprehensive link quality factor of S2, a hierarchical link maintenance mechanism is executed by comparing the predicted remaining lifetime with preset multi-level thresholds and taking into account the dynamic changes of the link quality factor. This aims to optimize resource utilization while ensuring communication continuity. Specifically, the mechanism includes the following sub-steps: S401: Preset Threshold and Initialization: When establishing a link, three remaining lifetime thresholds are pre-set based on the robot's motion characteristics, the performance of the communication equipment, and the reliability requirements of the application scenario: emergency threshold. First threshold Second threshold ,satisfy The threshold setting is based on the following criteria: Second threshold : It is usually set to be greater than the sum of the routing protocol convergence time and the power control response time, such as taking Seconds ensure sufficient time for observation before a minor risk is detected; First threshold : Set to a time greater than the minimum time required for route pre-switching, such as Second; Emergency threshold : Set to a time greater than the emergency processing time for data buffering and path switching after link interruption, such as Second; In addition, a percentage threshold for the decrease in the overall link quality factor is preset. and unit time window It is used to assist in the assessment of emergency risks.

[0051] S402: Obtain current link status parameters In each decision cycle, which is synchronized with the S1 acquisition cycle, the remaining lifetime of the link with neighboring robots is read from the stored results of S3. The comprehensive link quality factor is obtained from the real-time calculation results of S2. ;calculate In the most recent unit time window The rate of decline within:

[0052] in The value of t at the current time. unit time window If the denominator is zero or negative, the value is treated as zero.

[0053] S403: Hierarchical Decision-Making and Execution Will Compare with a preset threshold, and based on the comparison result and... Trigger the corresponding link hold operation: S41: Link stability maintenance ( ) When the remaining lifetime of the link is sufficient, it indicates that the current communication link is stable and there is no risk of immediate interruption. At this time, the system maintains the current link state and does not take any active adjustment measures. All data continues to be transmitted through the original path, the routing table remains unchanged, the communication power remains at the current level, and the status changes in subsequent periods are continuously monitored.

[0054] S42: Mild Risk Management ) When the remaining link lifetime enters the low-risk range, it indicates that the link has a potential deterioration trend but still has enough time to respond; the system determines it to be low-risk and triggers the power control mechanism to extend the link lifetime. Based on the current link margin, the local machine, such as And relative distance velocity, calculate the required transmit power increment This allows the increased power to compensate for the path loss caused by the increased distance. Construct a power control command frame, which includes the target neighboring robot identifier, the requested transmission power value, and the command validity timestamp, and send it to the peer robot through the current communication link; After receiving the instruction, the peer robot parses and adjusts the transmission power of its wireless communication module to the specified value, and replies with an acknowledgment frame; if no acknowledgment is received multiple times, the local robot can try to retransmit or gradually increase its own power; This machine will continue to monitor in subsequent cycles. Changes, if power adjustment is effective, It may rebound to That's all; if the situation continues to worsen, then proceed to the next level of treatment.

[0055] S43: Moderate Risk Management ) When the remaining lifetime of the link further decreases to the medium-risk range, it indicates that the link is about to be interrupted, but there is still a small amount of time to prepare. When it is determined to be of medium risk, the route pre-switching mechanism is triggered to establish a backup path in advance but not to immediately switch the primary path. The process of starting the backup path for the local robot is as follows: Based on the neighboring robot information table, one or more candidate paths from the local robot to the target neighboring robot are calculated using topology information (such as location and movement trend) or OLSR routing protocol; the path with the best overall quality is selected as the backup path. The backup path information is stored in the backup area of ​​the routing table, and probe packets are sent periodically to keep the path active, but data traffic is not switched to the backup path. At the same time, notify the robot routing management module to mark the current main link as "about to be interrupted" and prepare for switching; If during this period Restore to If the above conditions are met, the pre-switch status will be cancelled and the backup path will be cleared; if the situation continues to deteriorate, a switchover will be performed once an emergency risk level is reached.

[0056] S44: Emergency Risk Management ) When the remaining lifetime of a link is lower than the emergency threshold, and the overall link quality factor decreases by more than a preset percentage within a unit of time, it indicates that the link will be interrupted immediately. The system determines this as an emergency risk and performs immediate handover and data caching. Immediate Path Switching: Immediately switch the current primary path to the backup path established in S43; the specific operation is as follows: update the routing table, modify the link address of the target's neighboring robot to the link of the backup path, and cancel the original primary path's routing entry; at the same time, send update information to the link of the target's neighboring robot to avoid routing loops. Start data caching: During the switchover, the data to be sent by the upper layer application to the original link, as well as data that may have been sent but not yet acknowledged, are temporarily stored in the local data cache queue; the cache area is managed by a circular queue linked list, and the capacity is set according to the maximum expected interruption time; Retransmission of cached data: After the backup path is successfully established, data is read from the cache queue in order, retransmitted through the new path, and waits for confirmation from the other end. After successful confirmation, the data is cleared from the cache. If the new path also fails, the cache is continued and other paths are tried. Log and alarm recording: Record the time, cause, and amount of cached data of the switching event for subsequent analysis or maintenance.

[0057] S404: Status Rollback and Reset After the emergency risk management is completed, if the link returns to stability, such as rebounded to The following steps should be taken to gradually reset the hierarchical status: after confirming that all data has been successfully transmitted, clear the data cache; delete the backup path information and restore the monitoring mode to normal.

[0058] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, 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, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for maintaining ad hoc network links during robot movement, characterized in that, include: S1: Collect the physical layer parameters of the current communication link, and at the same time collect the motion state information of the local robot and nearby robots; The physical layer parameters include received signal strength indication, signal-to-noise ratio, and bit error rate; the motion state information includes velocity vector, acceleration, heading angle, and position coordinates. S2: Construct a signal quality sub-factor based on the physical layer parameters, construct a motion trend sub-factor based on the relative distance change rate and relative angle change rate in the motion state information, and perform weighted fusion of the signal quality sub-factor and the motion trend sub-factor to obtain a comprehensive link quality factor. The weighted fusion weight is dynamically adjusted according to the robot's current motion state. S3: Calculate the relative motion speed based on the position coordinates and velocity vectors of the local robot and neighboring robots, and predict the remaining link lifetime by combining the maximum communication distance of wireless communication; S4: Compare the remaining link lifetime with a preset first threshold, a second threshold, and an emergency threshold. If the emergency threshold is less than the first threshold, and the first threshold is less than the second threshold, then execute a tiered link maintenance mechanism. S41: When the remaining lifetime of the link is greater than or equal to the second threshold, maintain the current link state; S42: When the remaining lifetime of the link is between the first threshold and the second threshold, it is determined to be a mild risk. A power control command is sent to the peer robot to perform power control in order to maintain the link. S43: When the remaining lifetime of the link is between the emergency threshold and the first threshold, it is determined to be of medium risk, triggering route pre-switching, establishing a backup path but not immediately switching the primary path; S44: When the remaining lifetime of the link is lower than the emergency threshold and the overall link quality factor decreases by more than a preset percentage per unit time, it is determined to be an emergency risk, and the system immediately switches to the backup path and starts data caching.

2. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, S1 specifically includes: S11: Trigger the data acquisition task with a preset period T, or trigger real-time acquisition when the communication module detects a link event, and use a distributed clock synchronization protocol to give the data collected by the local robot and neighboring robots a unified time tag. S12: Read real-time physical layer data through the underlying driver interface of the wireless communication module; obtain the received signal strength indicator RSSI by reading the network card register; read the signal-to-noise ratio SNR from the underlying status register of the communication baseband chip; calculate the bit error rate BER by statistically analyzing the proportion of physical layer decoding error bits to the total number of bits. S13: Collects the robot's motion state information through the sensor bus and robot operating system node; reads the inertial measurement unit data and obtains the acceleration after fusion by Kalman filtering; reads the wheel encoder calculation results to obtain the velocity vector; reads the attitude calculation results after fusion of electronic compass and inertial measurement unit data to obtain the heading angle; obtains latitude and longitude coordinates through satellite navigation and converts them into local position coordinates; S14: Send a status request command to the neighboring robot through the wireless communication link, receive the motion status information returned by the neighboring robot, and distinguish different neighboring robots according to the source address in the data frame, and store them in the local dynamic neighbor information table. S15: The physical layer parameters are processed using a sliding window average filtering or first-order low-pass filtering algorithm, and the motion state data caused by the mismatch of timestamps of neighboring robots due to network latency is corrected using an interpolation algorithm.

3. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, The construction of the signal quality sub-factor in S2 specifically includes: The received signal strength indication is normalized to obtain And Cut off to interval; The signal-to-noise ratio is normalized to obtain And Cut off to The bit error rate is normalized using a negative exponential function within a certain range. The three normalization factors are then combined according to preset weights to obtain the signal quality sub-factor. .

4. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, The construction of the motion trend sub-factor in S2 specifically includes: Calculate the rate of change of relative distance Calculate the rate of change of relative angle. Mapping the rate of change of relative distance to a stability component Mapping the relative angle change rate to a stability component ,based on and The motion trend sub-factors were obtained by comprehensive analysis. .

5. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, The dynamic adjustment of the weighted fusion weights in S2 specifically includes: Define the intensity index of exercise And based on the intensity of exercise index Determine the signal quality factor weight and movement trend factors weight The signal quality factor and motion trend factor are weighted and fused to obtain the comprehensive link quality factor. .

6. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, S3 specifically includes: S31: Extract the current location coordinates of the local machine at time t from the dynamic neighbor information table. and velocity vector and the location coordinates of nearby robots and velocity vector ; S32: Calculate the relative motion parameters, including the relative position vector. Current Euclidean distance Relative velocity vector and relative radial velocity ; S33: Read the system's preset maximum effective wireless communication range. ; S34: Predicting the remaining link lifetime using the linear motion assumption ; S35: The calculated result The corresponding neighboring robot identifier is associated with the storage and used as an input parameter for subsequent hierarchical link maintenance mechanisms.

7. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, The preset threshold setting method in S4 includes: Second threshold Set to be greater than the sum of the routing protocol convergence time and the power control response time; First threshold Set to a time greater than the minimum time required for route pre-switching; Emergency threshold Set to a time greater than the emergency processing time for data buffering and path switching after link interruption.

8. The method for maintaining ad-hoc network links during robot movement according to claim 1, characterized in that, The S42 mild risk handling specifically includes: This machine is based on the current link margin Given the relative velocity, calculate the required increase in transmit power; Construct a power control command frame containing the identifier of the target's neighboring robot, the requested transmission power value, and the command's valid timestamp, and send it to the peer robot through the current communication link; After receiving the instruction, the peer robot adjusts the transmission power of its wireless communication module to the specified value and replies with an acknowledgment frame; This machine will continue to monitor in subsequent cycles. Changes, if power adjustment is effective and rebounded to If the above conditions are met, the current state will be maintained; if the situation continues to worsen, the process will proceed to the next level, S43.

9. A method for maintaining ad-hoc network links during robot movement according to claim 8, characterized in that, The S43 moderate risk management specifically includes: The machine initiates a backup path, calculates candidate paths from the machine to the target neighboring robot based on the neighboring robot information table and routing protocol, and selects the path with the best overall quality as the backup path; The backup path information is stored in the backup area of ​​the routing table, and probe packets are sent periodically to keep the backup path active, but data traffic is not switched to the backup path. The robot routes the current main link to indicate that it is about to be interrupted. like Restore to If the above conditions are met, the pre-switch status will be cancelled and the backup path will be cleared; if the situation continues to deteriorate, the switchover will be performed in the S44 emergency risk process.

10. A method for maintaining ad-hoc network links during robot movement according to claim 9, characterized in that, The S44 emergency risk management specifically includes: Calculate the overall link quality factor within a unit time window The decline within ;when If the situation is deemed an emergency risk, the current primary path is immediately switched to the established backup path, the routing table is updated, and the original primary path's routing entries are removed. During the switchover, the data to be sent by the upper-layer application to the original link, as well as the data that has been sent but not yet acknowledged, are temporarily stored in the local data cache queue; After the backup path is successfully established, data is read from the cache queue in sequence, resent through the new path, and waits for confirmation from the peer robot. The time, reason and amount of cached data of the switching event are recorded. After the emergency risk management is completed, if rebounded to If the data cache is cleared, the backup path information is deleted, and the system returns to normal monitoring mode.

Images