An automatic guided vehicle path planning method based on swarm intelligence

Abstract

The present application relates to the technical field of intelligent path planning, and particularly relates to an automatic guided vehicle path planning method based on swarm intelligence. The method comprises the following steps: obtaining multiple vehicle position information based on swarm intelligence, identifying candidate paths and drivable space channels, judging low-speed stable state in combination with driving speed and space congestion degree, reselecting stable channels and allocating residence time, and realizing dynamic path conversion and optimized scheduling by comprehensively considering channel superiority and residence time. The present application realizes the collaborative optimization of multi-vehicle path planning by introducing swarm intelligence, multi-dimensional dynamic perception, low-speed stable state detection, stable space channel screening, and smooth constraint path conversion and short-time obstacle avoidance mechanism, which significantly improves the overall transportation efficiency and operation reliability in terms of congestion management, channel safety, path switching smoothness and orderly traffic of multiple vehicles.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent path planning, in particular to an automatic guided vehicle path planning method based on swarm intelligence. BACKGROUND

[0002] The existing automatic guided vehicle path planning usually relies on static maps or preset routes, and the vehicle travels along the fixed path, lacking real-time adaptability to dynamic environment. When multiple vehicles share the channel, the simple rules such as "first come first served" or queue-based are often used to handle intersections and congestion, which cannot quantify the degree of channel congestion and cannot make dynamic judgments combined with vehicle speed, spatial distribution and real-time state. At the same time, the recognition of low-speed stable state is insufficient, and it is often only judged by a single speed threshold, which is easy to misjudge the congestion area, thereby affecting the scheduling efficiency. The path conversion and short-time obstacle avoidance strategy is mostly simple stopping, waiting or detouring, lacking smooth constraint and dynamic optimization control, resulting in frequent start-stop of vehicles, non-smooth path and possible impact on cargo safety. In addition, the channel superiority evaluation lacks multi-factor comprehensive judgment, usually only considering a single index such as speed or distance, resulting in lack of scientific basis for path selection and easy formation of bottleneck channel congestion. SUMMARY

[0003] Therefore, it is necessary to provide an automatic guided vehicle path planning method based on swarm intelligence to solve at least one of the above technical problems.

[0004] To achieve the above purpose, an automatic guided vehicle path planning method based on swarm intelligence, the method comprising the following steps:

[0005] Step S1: confirming a plurality of candidate paths based on the position information between the plurality of automatic guided vehicles and the target; confirming the space channel that the vehicle can travel along according to the center line and the width of the candidate path;

[0006] Step S2: acquiring the running speed of the automatic guided vehicle and the congestion degree between the vehicle and the adjacent vehicle; detecting whether the vehicle is in a low-speed stable state through the running speed and the congestion degree;

[0007] Step S3: screening the stable space channel in the space channel according to the low-speed stable state, and setting the residence time based on the stable space channel; calculating the channel superiority degree of the stable space channel, and confirming whether the vehicle needs to convert the path combined with the residence time.

[0008] Preferably, step S2 comprises the following steps:

[0009] Step S21: acquiring the running speed of the automatic guided vehicle through the vehicle-mounted motion sensor;

[0010] Step S22: enabling the fleet position interconnection function on the automated guided vehicle, receiving real-time coordinate information periodically broadcast by adjacent vehicles, wherein the real-time coordinate information includes the planar position of the vehicle in the coordinate system of the working scene;

[0011] Step S23: for any automated guided vehicle, calculating the straight-line distance between the vehicle and the adjacent vehicle according to the position information and the received position information of the adjacent vehicle, and confirming the relative motion direction of the automated guided vehicle in combination with the change trend of the planar position;

[0012] Step S24: confirming the degree of spatial congestion of the vehicle and the adjacent vehicle in the same spatial channel according to the relative motion direction; and detecting whether the vehicle is in a low-speed stable state according to the driving speed and the degree of congestion.

[0013] Preferably, step S24 comprises the following steps:

[0014] Step S241: determining whether the vehicle and the adjacent vehicle are in the same spatial channel according to the relative motion direction;

[0015] Step S242: after determining that the vehicle and the adjacent vehicle are in the same spatial channel, evaluating the degree of spatial congestion of the channel according to the distance between the vehicle and the adjacent vehicle, the lateral gap, and the front and rear distribution;

[0016] Step S243: numerically quantifying the degree of spatial congestion of the channel based on a preset spatial congestion quantification table to obtain a spatial congestion degree coefficient; and detecting whether the vehicle is in a low-speed stable state using the spatial congestion degree coefficient and the driving speed.

[0017] Preferably, step S243 comprises:

[0018] When the driving speed is less than a preset speed threshold and the spatial congestion degree coefficient is greater than or equal to a preset congestion threshold, the vehicle triggers a low-speed stable mode.

[0019] Preferably, the construction of the spatial congestion quantification table in step S243 comprises:

[0020] Collecting vehicle position distribution data under different operating scenarios, wherein the vehicle position distribution data includes longitudinal distance intervals and lateral gap intervals;

[0021] Dividing the longitudinal distance intervals and the lateral gap intervals of the spatial channel into several levels, and defining a congestion degree identifier for each level;

[0022] For each congestion level, the passing speed, brake frequency and waiting time of the trolley collected in actual operation are analyzed, the numerical coefficients corresponding to each congestion level are extracted, and a space congestion quantification table is obtained.

[0023] Preferably, step S3 comprises the following steps:

[0024] Step S31: In the detected low-speed stable state of the trolley, the space channel where the trolley is located is screened out and marked as a candidate stable space channel;

[0025] Step S32: Vehicle operation anomaly detection is performed on each candidate stable space channel to confirm the stable space channel, and a residence time is allocated to each stable space channel.

[0026] Step S33: The channel superiority degree of each stable space channel is calculated, and a comprehensive comparison is made in combination with the residence time. When the channel superiority degree is less than a preset superiority degree threshold value, and the residence time is greater than or equal to a preset time threshold value, a path conversion suggestion signal is generated and sent to the trolley dispatching control module.

[0027] Step S34: When the channel superiority degree is greater than or equal to the preset superiority degree threshold value, or the residence time is less than the preset time threshold value, the original channel is kept running.

[0028] Preferably, the vehicle operation anomaly detection on each candidate stable space channel in step S32 comprises:

[0029] The running data of the trolley in each candidate stable space channel is collected, including the vehicle speed, acceleration, obstacle distance in the channel and vehicle lateral offset amount;

[0030] When the vehicle speed deviates from the set low-speed interval by more than ±0.2 m / s, or the absolute value of the acceleration continuously exceeds 0.4 m / s² for more than 3 seconds, or the obstacle distance is less than 1.0 m and the duration is more than 2 seconds, or the vehicle lateral offset amount exceeds 10 cm, the vehicle is determined to be running abnormally;

[0031] In a continuous monitoring period, if no vehicle operation anomaly is detected, the space channel is marked as a stable space channel; if any anomaly is detected in the monitoring period, the space channel is marked as an unstable space channel and is excluded, wherein the continuous monitoring period is 30-120 s.

[0032] Preferably, the formula for calculating the channel superiority degree of each stable space channel in step S33 is as follows:

[0033] ;

[0034] In the formula, is a channel superiority degree coefficient, is a normalized average speed, is a normalized channel congestion coefficient, is a normalized traffic efficiency, is a normalized stability coefficient, is a normalized risk degree, , ... is a weight coefficient, and satisfies + + + + =1.

[0035] Preferably, when the path conversion suggestion signal is received by the transport vehicle scheduling control module of the transport vehicle, the method further comprises:

[0036] setting the space channel where the transport vehicle is located as a first channel;

[0037] confirming the converted space channel based on the path conversion suggestion signal and setting it as a second channel;

[0038] performing transport vehicle tracking in the second channel using smooth constraint predictive control and returning to the first channel after completing short-time obstacle avoidance in the second channel, wherein the smooth constraint is specifically trajectory curvature and its change rate, and longitudinal acceleration and its change rate constraint.

[0039] Preferably, the short-time obstacle avoidance further comprises:

[0040] when it is detected that the transport vehicle intersects with a transport vehicle in the opposite direction, the transport vehicles are assigned traffic priority to perform vehicle sequential traffic operation.

[0041] The present application has the following advantages:

[0042] I. By introducing the idea of swarm intelligence, the position information, driving speed and space congestion degree between multiple automatic guided transport vehicles and the target are comprehensively utilized to realize dynamic perception and collaborative decision-making among multiple vehicles in the path planning process, which can effectively solve the problems of path conflict, congestion backlog and scheduling imbalance in the prior art, and significantly improve the overall transportation efficiency. Through the identification of the drivable space channel based on the center line and width of the candidate path, it is ensured that the path planning has higher spatial matching accuracy and feasibility.

[0043] II. In low-speed steady-state detection, the vehicle-mounted motion sensor is used in conjunction with the vehicle fleet position interconnection function, combined with the real-time coordinate information of the adjacent transport vehicles, to accurately calculate the straight-line distance between the two vehicles and the relative motion direction, and to quantitatively evaluate through the space congestion quantization table, so that the congestion degree judgment is more objective and quantifiable, ensuring the reliability of low-speed steady-state recognition. At the same time, through the hierarchical design of longitudinal distance interval and lateral gap interval, as well as the correlation analysis with passing speed, brake frequency and waiting time, precise congestion quantification in different scenarios is realized, with stronger adaptability.

[0044] III. In the stable space channel screening process, an abnormal vehicle operation detection mechanism is introduced to comprehensively monitor multiple-dimensional operation parameters such as vehicle speed, acceleration, obstacle distance and vehicle lateral deviation, which can timely eliminate unstable channels and ensure the safety and continuity of path selection. Through comprehensive judgment of residence time and channel superiority, frequent and ineffective path switching is avoided, improving the stability of operation and the rationality of scheduling.

[0045] IV. In the path conversion execution link, the transport vehicle is tracked through smooth constraint predictive control, which constrains the trajectory curvature and its rate of change, longitudinal acceleration and its rate of change, ensuring the smoothness of the path switching process and the safety of vehicle control. At the same time, in the short-time obstacle avoidance process, a passing priority distribution mechanism is introduced, which effectively solves the problem of priority passing when relative direction transport vehicles intersect, avoids the efficiency reduction caused by mutual yielding of multiple vehicles, and realizes orderly passing under the cooperation of multiple vehicles. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 A schematic diagram of the step flow of the automatic guided vehicle path planning method based on swarm intelligence is shown.

[0047] Figure 2 A detailed implementation step flow diagram of step S2 in the method is shown. Figure 1 A detailed implementation step flow diagram of step S2 in the method is shown.

[0048] Figure 3 A transport schematic diagram of the automatic guided vehicle is shown.

[0049] The implementation of the present application, functional features and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0050] The technical method of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0051] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0052] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0053] To achieve the above objectives, please refer to Figures 1 to 3 A path planning method for automated guided vehicles (AGVs) based on swarm intelligence, the method comprising the following steps:

[0054] Step S1: Based on the positional information between multiple automated guided vehicles and the target, identify several candidate paths; determine the space channel through which the vehicles can travel based on the centerline and width of the candidate paths;

[0055] In one embodiment, reference may be made to Figure 3 The system receives real-time position information between multiple automated guided vehicles (AGVs) and the target. For example, the position information acquired at the first moment can be defined as the first position data, where the first moment is the start time of the current scheduling cycle. Based on the first position data and obstacle information in the environmental map, the system can calculate several candidate paths from the current position of each vehicle to the target position.

[0056] In some embodiments, candidate paths can be used to characterize the drivable route of a transport vehicle from its current location to a target location. For example, the system can generate a drivable spatial passage by analyzing the geometry of the candidate path, extracting the path's centerline, and determining the total width of the passage based on the transport vehicle's own width and lateral safety distance.

[0057] For example, if the minimum safe distance from obstacles is met on both sides of the centerline of a candidate path throughout the entire path, the spatial passage of the path can be determined as a valid passage; if the distance between a certain section of the centerline and the obstacle is insufficient, the passage can be determined as an unusable passage and eliminated.

[0058] For example, if the spatial channel of the same path is detected to be unblocked and meet the width and safety distance requirements at the first time and the second time, the channel can be determined as a stable drivable channel. That is, the channel with higher priority can be screened by judging the continuity and stability of the channel, so as to facilitate subsequent accurate path tracking and obstacle avoidance operations.

[0059] Step S2: Obtain the driving speed of the automated guided vehicle and the congestion degree between the vehicle and the adjacent vehicle; and detect whether the vehicle is in a low-speed stable state through the driving speed and the congestion degree;

[0060] In an embodiment, the driving speed information of the automated guided vehicle and the congestion degree data between the vehicle and the adjacent vehicle can be received in real time. For example, the driving speed obtained at the first time can be defined as first speed data, and the congestion degree obtained at the first time can be defined as first congestion data, and the first time is the starting time of the current detection period. The system can judge the running state of the vehicle at the first time based on the first speed data and the first congestion data.

[0061] In another embodiment, the low-speed stable state can be used to represent that the vehicle has a low speed and small fluctuation during driving, and the surrounding environment is in a relatively stable traffic state. For example, when the driving speed of the vehicle is lower than a preset low-speed threshold and the speed change rate is less than a stable threshold, and the average distance between the vehicle and the adjacent vehicle is greater than a safety distance, it can be determined that the vehicle is in a low-speed stable state.

[0062] For example, when the driving speed of the vehicle is detected to be lower than the low-speed threshold, the speed change rate is lower than the stable threshold, and the congestion degree of the adjacent vehicle does not exceed the preset congestion threshold at the first time and the second time, it can be determined that the vehicle continues to be in a low-speed stable state.

[0063] For example, by continuously monitoring the speed and congestion degree change trend of the vehicle at multiple continuous times, it can also be judged whether the vehicle transits from the low-speed stable state to the congestion state or the acceleration state, so as to facilitate the subsequent scheduling system to take path optimization or vehicle distance adjustment measures in advance.

[0064] Step S3: Screen the stable spatial channel in the spatial channel according to the low-speed stable state, and set a residence time based on the stable spatial channel; calculate the channel superiority degree of the stable spatial channel, and confirm whether the vehicle needs to perform path conversion in combination with the residence time.

[0065] In an embodiment, based on the low-speed steady state detected in step S2, a channel that remains unblocked, has a width and a safety distance meeting the requirements, and has no potential conflicts at multiple consecutive detection moments can be selected from the set of spatial channels as a stable spatial channel. For example, when a channel does not appear to have an obstacle intrusion or insufficient width at two consecutive detections, the channel can be determined to be a stable spatial channel.

[0066] In some embodiments, a dwell time can be set for a stable spatial channel, representing the time for which the trolley remains driving in the channel without switching to other channels. For example, the dwell time can be counted from the time when the trolley enters the stable spatial channel, and ends when the channel state changes or the trolley performs path switching.

[0067] For another example, the channel superiority degree can be used to reflect the comprehensive advantages of a stable spatial channel in terms of traffic efficiency and safety. The system can calculate the superiority degree score according to factors such as channel length, curvature, minimum safety distance from obstacles, expected travel time, and potential conflict risk, and sort the stable spatial channels.

[0068] For another example, when the superiority degree of a stable spatial channel is lower than that of other available channels, and the dwell time has exceeded a preset dwell threshold, it can be determined that the trolley needs to perform path switching to enter a channel with a higher superiority degree; when the difference in superiority degree is insufficient or the dwell time does not exceed the threshold, the current channel can be maintained to reduce the running fluctuations caused by frequent switching.

[0069] As an example of the present application, refer to FIG. 1, which shows a system for guiding a trolley in a warehouse according to an embodiment of the present application. In this example, the step S2 includes: Figure 2

[0070] Step S21: Obtain the driving speed of the automated guided trolley through the vehicle-mounted motion sensor;

[0071] Step S22: Enable the vehicle fleet position interconnection function on the automated guided trolley, and receive real-time coordinate information periodically broadcast by adjacent trolleys, wherein the real-time coordinate information includes the planar position of the trolley in the work scene coordinate system;

[0072] Step S23: For any automated guided trolley, calculate the straight-line distance between the two trolleys according to the position information and the position information of the adjacent trolley received, and confirm the relative motion direction of the automated guided trolley in combination with the trend of the planar position change;

[0073] Step S24: Confirm the degree of spatial congestion of the trolley and the adjacent trolley in the same spatial channel according to the relative motion direction; and detect whether the trolley is in a low-speed steady state according to the driving speed and the degree of congestion.

[0074] ​In an embodiment, the on-board motion sensor can include an inertial measurement unit, a wheel speed sensor, an odometer, etc. to detect the driving speed of the truck in real time. The system can record the speed data at a preset sampling period (e.g. once per second) and transmit it to the dispatch control system for subsequent state determination. The truck fleet position interconnection function can be implemented based on a wireless communication module (e.g. industrial wireless local area network, ultra-wideband communication or dedicated short-range communication), and adjacent trucks will send their two-dimensional position coordinates in the scene coordinate system at a fixed broadcast period (e.g. 500 milliseconds) to achieve real-time position sharing between trucks.

[0075] In some embodiments, the system can use the Euclidean distance formula to calculate the straight-line distance between the truck and the adjacent truck, and determine the relative motion direction by the direction of coordinate change in a plurality of consecutive sampling periods, such as approaching each other, moving away from each other or keeping parallel. The degree of space congestion can be evaluated by the number of adjacent vehicles within a unit distance range and the relative motion direction. For example, in the same space channel, if the truck speed is low and the number of adjacent trucks around is large, and the relative motion direction tends to be the same, it can be determined to be in a low-speed stable state; otherwise, it is in a congested or dispersed driving state.

[0076] In another embodiment, the low-speed threshold can be set to 0.3 to 0.8 meters per second, the stable speed change rate threshold can be set to 0.05 to 0.1 m / s², and the space congestion degree determination distance range can be set to 2 to 5 meters. The reason for setting the low-speed threshold in the above range is that for most small and medium-sized automatic guided trucks, the speed range can keep stable driving without affecting the work efficiency; below 0.3 meters per second, the work efficiency will be significantly reduced, and above 0.8 meters per second, frequent acceleration and deceleration will occur between vehicles. The range of the stable speed change rate threshold is set to ensure stable driving at low speed and not to cause channel blockage due to frequent acceleration or deceleration. The distance range of the degree of space congestion is determined in combination with the length of the truck (usually 1 to 1.5 meters) and the necessary safety distance, and the range of 2 to 5 meters can reflect the actual density between vehicles and avoid misjudgment of vehicles in adjacent but different channels as congested.

[0077] Preferably, step S24 comprises the following steps:

[0078] Step S241: determining whether the truck and the adjacent truck are in the same space channel according to the relative motion direction;

[0079] Step S242: after determining that they are in the same space channel, evaluating the degree of space congestion of the channel according to the distance between the truck and the adjacent truck, the lateral gap and the front and rear distribution;

[0080] Step S243: Numerically quantifying the degree of space congestion of the channel based on the preset space congestion quantification table, to obtain a space congestion degree coefficient; and determining whether the tugger is in a low-speed stable state by using the space congestion degree coefficient and the driving speed.

[0081] In an embodiment, the relative motion direction of the two vehicles can be determined according to the change trend of the planar position of the tugger and the adjacent tugger in the work scene coordinate system, in combination with the direction of the channel center line. When the relative motion direction is consistent and the lateral offset of the two vehicles is less than a preset threshold, it can be determined that the two vehicles are in the same space channel; otherwise, it is determined that they are in different channels. The system can calculate the straight-line distance between the tugger and the adjacent tugger in front and behind, and comprehensively evaluate the congestion degree in combination with the lateral gap between the two vehicles. For example, when the front distance is relatively short and the lateral gap is insufficient, the system can determine that the region is a high-congestion region; otherwise, it is determined to be a low-congestion region.

[0082] In some embodiments, the space congestion quantification table can be graded according to the congestion degree, for example, the coefficient is divided into continuous values between 0 and 1, and the larger the coefficient, the more congested the channel. The system can jointly determine the coefficient and the current driving speed of the tugger: when the coefficient is greater than a set congestion threshold and the driving speed is lower than a low-speed threshold, it can be determined that the tugger is in a low-speed stable state.

[0083] In another embodiment, the lateral offset threshold can be set to 0.3 to 0.6 meters, the front and rear distance determination range can be set to 1.5 to 4 meters, the congestion threshold of the space congestion degree coefficient can be set to 0.6 to 0.8, and the low-speed threshold can be set to 0.3 to 0.8 meters per second. The lateral offset threshold is set to 0.3 to 0.6 meters, considering that the width of most automatic guided tugger is about 0.8 to 1 meter, which can effectively distinguish whether it is in the same channel; the determination range of the front and rear distance combines the length of the tugger (usually 1 to 1.5 meters) and the requirement of safety distance, which can cover the passing interval in common work scenes. The congestion threshold of the space congestion degree coefficient is set to 0.6 to 0.8, in order to timely trigger the low-speed stable state determination in medium and high congestion situations, and avoid frequent braking of the vehicle due to congestion; the low-speed threshold is set to 0.3 to 0.8 meters per second, which takes into account the work efficiency while ensuring smooth driving.

[0084] Preferably, the step S243 of detecting whether the tugger is in a low-speed stable state by using the space congestion degree coefficient and the driving speed comprises:

[0085] When the driving speed is less than a preset speed threshold and the space congestion degree coefficient is greater than or equal to a preset congestion threshold, the tugger triggers a low-speed stable mode.

[0086] In an embodiment, the tugger triggers the low-speed stabilization mode when the travel speed is less than a preset speed threshold and the space congestion degree coefficient is greater than or equal to a preset congestion threshold. In the low-speed stabilization mode, the tugger travels at a low speed smoothly while maintaining a safe distance from the adjacent tugger to ensure the safety and stability of travel in the aisle.

[0087] In some embodiments, the system can periodically monitor the travel speed and the space congestion degree coefficient of the tugger, and confirm that the tugger enters the low-speed stabilization mode when the speed threshold and the congestion threshold conditions are met for two or more consecutive detection periods to avoid misjudgment of the state due to transient fluctuations.

[0088] For example, in some embodiments, the speed threshold can be set to 0.3-0.8 m / s, and the congestion threshold of the space congestion degree coefficient can be set to 0.6-0.8. The speed threshold is set to 0.3-0.8 m / s to ensure that the tugger can still maintain stable travel at low speed without affecting the work efficiency; below 0.3 m / s leads to too low work efficiency, and above 0.8 m / s poses a safety hazard in a crowded aisle. The congestion threshold of the space congestion degree coefficient is set to 0.6-0.8 to trigger the low-speed stabilization mode in time in a medium-high congestion state, ensuring that the tugger can run smoothly in the aisle and avoiding frequent braking or collision risks.

[0089] Preferably, the construction of the space congestion quantification table in step S243 specifically includes:

[0090] Collecting tugger position distribution data under different running scenarios, wherein the tugger position distribution data includes longitudinal distance intervals and lateral gap intervals;

[0091] Dividing the longitudinal distance intervals and the lateral gap intervals of the space aisle into several levels, and defining a congestion degree identifier for each level;

[0092] For each congestion degree identifier, performing correlation analysis on the travel speed, brake frequency, and waiting time of the tugger collected in actual operation, extracting the numerical coefficient corresponding to each congestion level, and obtaining the space congestion quantification table.

[0093] In an embodiment, the tugger position distribution data under different running scenarios is collected, wherein the tugger position distribution data includes longitudinal distance intervals and lateral gap intervals. For example, the actual positions of the tugger are collected under different scenarios such as multiple space aisles and intersection points, curves, and straight lines, and the distance information of each tugger from the front, rear, and both sides is recorded.

[0094] Based on the longitudinal distance interval and the lateral gap interval of the space channel, the channel is divided into several levels, and a congestion degree identifier is defined for each level. For example, the longitudinal distance interval can be divided into three levels of low, medium and high, and the lateral gap interval can be divided into three levels of small, medium and large, thereby forming different congestion combinations, each corresponding to a congestion degree identifier.

[0095] For each congestion degree identifier, the forklift passing speed, brake frequency and waiting time collected in actual operation are analyzed, the numerical coefficients corresponding to each congestion level are extracted, and a space congestion quantification table is obtained. By analyzing the running characteristics of the forklift under different congestion levels, the congestion degree of the channel can be quantified, providing a basis for subsequent low-speed stable state determination.

[0096] In some embodiments, the longitudinal distance interval can be set to 1-4 meters, and the lateral gap interval can be set to 0.3-0.8 meters. The congestion degree identifier can be divided into three levels of low, medium and high, corresponding to space congestion coefficients of 0.2-0.4, 0.4-0.6 and 0.6-0.8 respectively. The reason for setting the longitudinal distance interval to 1-4 meters is based on the actual situation that the length of the forklift is usually 1-1.5 meters, to ensure that the longitudinal distance can reflect the vehicle density and cover common operating scenarios; the lateral gap interval of 0.3-0.8 meters is determined in combination with the width of the forklift and the safety lateral distance, which can distinguish different congestion levels. The congestion degree coefficient is set to the range of 0.2-0.8, in order to quantify the congestion state of the space channel, so that the system can reasonably determine the low-speed stable mode according to different congestion coefficients, while considering the driving efficiency and safety.

[0097] Preferably, step S3 comprises the following steps:

[0098] Step S31: In the detected forklift low-speed stable state, the space channel where the forklift is located is selected and marked as a candidate stable space channel;

[0099] Step S32: Vehicle running anomaly detection is performed on each candidate stable space channel to confirm the stable space channel; and a residence time is allocated to each stable space channel;

[0100] Step S33: The channel superiority degree of each stable space channel is calculated, and a comprehensive comparison is made in combination with the residence time; when the channel superiority degree is less than a preset superiority degree threshold and the residence time is greater than or equal to a preset time threshold, a path conversion suggestion signal is generated and sent to the forklift dispatching control module;

[0101] Step S34: When the channel superiority degree is greater than or equal to the preset superiority degree threshold, or the residence time is less than the preset time threshold, the original channel is kept running.

[0102] In an embodiment, based on the low-speed steady state detected in step S2, the system can filter out the channels from the set of spatial channels that have remained open, with width and safety distance meeting the requirements, in multiple consecutive detection periods, and mark them as candidate stable spatial channels. The running anomaly detection can include judging the abnormal behaviors of the trolley in the channel, such as acceleration, deceleration, stopping, lateral deviation, etc. After confirming the stability of the channel, the system can start timing from the moment the trolley enters the channel, calculate the residence time of the trolley in the channel, and use it for subsequent channel superiority degree and path switching judgment.

[0103] In some embodiments, the channel superiority degree can be calculated by comprehensively considering factors such as channel length, curvature, minimum safety distance of obstacles, estimated travel time, and potential conflict risk. When the channel superiority degree is low and the trolley stays in the channel for too long, the system generates a path switching suggestion signal to prompt the trolley dispatch system to adjust the travel path. When the channel superiority degree is high or the trolley stays in the channel for a short time, the system maintains the trolley traveling along the original channel, avoiding frequent path switching, and improving overall travel efficiency and stability.

[0104] In another embodiment, the channel superiority degree threshold can be set to 0.6 to 0.8, and the residence time threshold can be set to 10 to 30 seconds. The reason for setting the channel superiority degree threshold to 0.6 to 0.8 is based on the comprehensive evaluation of trolley channel travel efficiency and safety, to ensure that path switching is triggered when the channel conditions are poor, improving travel safety and efficiency. The residence time threshold is set to 10 to 30 seconds, taking into account the fact that the trolley may stay in the channel due to traffic congestion or work factors in the low-speed steady state. Exceeding the threshold indicates that the channel has low efficiency or potential congestion, thus reasonably triggering path optimization.

[0105] Preferably, the vehicle running anomaly detection for each candidate stable spatial channel in step S32 includes:

[0106] Collecting the running data of the trolley in each candidate stable spatial channel, including vehicle speed, acceleration, obstacle distance in the channel, and vehicle lateral deviation amount;

[0107] When the vehicle speed deviates from the set low-speed interval by more than ±0.2 m / s, or the absolute value of the acceleration exceeds 0.4 m / s² continuously for more than 3 seconds, or the obstacle distance is less than 1.0 m and the duration is more than 2 seconds, or the vehicle lateral deviation amount exceeds 10 cm, the vehicle is determined to be running abnormally;

[0108] In the consecutive monitoring period, if no vehicle running anomaly is detected, the spatial channel is marked as a stable spatial channel; if any anomaly is detected in the monitoring period, the spatial channel is marked as an unstable spatial channel and is excluded, wherein the consecutive monitoring period is 30-120 seconds.

[0109] In an embodiment, the running data of each candidate stable space channel is collected, including vehicle speed, acceleration, obstacle distance in the channel, and vehicle lateral offset. In a continuous monitoring period, the system records and analyzes the running state of the vehicle in real time to determine the running stability of the channel. When the vehicle speed deviates from the set low speed interval by more than ±0.2 m / s, or the absolute value of the acceleration continuously exceeds 0.4 m / s2and the duration exceeds 3 seconds, or the obstacle distance is less than 1.0 m and the duration exceeds 2 seconds, or the vehicle lateral offset exceeds 10 cm, the system determines that the vehicle has a running anomaly.

[0110] In a continuous monitoring period, if no running anomaly of the vehicle is detected, the space channel is marked as a stable space channel; if any anomaly is detected in the monitoring period, the space channel is marked as an unstable space channel and is excluded. The continuous monitoring period can be set to 30 to 120 seconds to balance the monitoring accuracy and response timeliness.

[0111] In some embodiments, through real-time analysis of various running parameters in a continuous monitoring period, the system can ensure that only when the channel running condition is continuously stable, the channel is included in the stable space channel set, thereby improving the safety and reliability of path selection and scheduling.

[0112] Preferably, the formula for calculating the channel superiority degree of each stable space channel in step S33 is as follows:

[0113] ;

[0114] In the formula, is the channel superiority degree coefficient, is the normalized average speed, is the normalized channel congestion coefficient, is the normalized traffic efficiency, is the normalized stability coefficient, is the normalized risk degree, , ... is the weight coefficient, and satisfies + + + + =1.

[0115] In an embodiment, the parameter settings are as follows: average speed reflects the smoothness of the channel, the higher the speed, the greater the channel superiority degree; channel congestion coefficient reflects the density of vehicles in the channel, the higher the congestion, the lower the superiority degree, so 1- ; passage efficiency represents the efficiency of the trolley actually passing through the passage, including factors such as stasis in the passage, brake times, etc., the higher the efficiency, the higher the degree of superiority; stability coefficient for measuring the stability of the trolley running in the passage, such as low-speed stable state duration and abnormal times, the higher the stability, the higher the degree of superiority; risk degree represents the potential safety hazards of the passage, such as insufficient minimum distance from obstacles or collision probability, the higher the risk, the lower the degree of superiority, so it is taken as a negative term in the formula; weight coefficient According to the actual operation needs and safety strategies of the trolley, the weight coefficient can be configured to balance the influence of speed, congestion, efficiency, stability and risk on the superiority of the passage.

[0116] The passage selection in the prior art usually only considers a single factor (such as speed or passage length), which easily ignores the congestion level and safety risk in the passage, resulting in frequent braking of the trolley or potential collisions. The formula comprehensively evaluates the superiority of the passage in multiple dimensions, taking into account speed, congestion, efficiency, stability and risk, which can realize the quantification and sorting of the running state of the passage, make the path selection more scientific and reasonable, reduce the risk of frequent path switching and accidents, and improve the overall operation efficiency and safety of the trolley.

[0117] Preferably, when the trolley dispatching control module of the trolley receives the path switching suggestion signal, it further comprises:

[0118] setting the space passage where the trolley is located as the first passage;

[0119] confirming the converted space passage based on the path switching suggestion signal and setting it as the second passage;

[0120] using smooth constraint predictive control for trolley tracking in the second passage and returning to the first passage after completing short-term obstacle avoidance in the second passage, wherein the smooth constraint is specifically trajectory curvature and its rate of change, and longitudinal acceleration and its rate of change constraint.

[0121] In an embodiment, the space passage where the trolley is located is set as the first passage, which is used to record the information of the current driving passage, so as to realize the return after path switching. Based on the path switching suggestion signal, the converted space passage is confirmed and set as the second passage, which is used for temporary switching of the driving path of the trolley to realize passage optimization. Smooth constraint predictive control is used for trolley tracking in the second passage, and the trolley returns to the first passage after completing short-term obstacle avoidance in the second passage. The smooth constraint includes trajectory curvature and its rate of change constraint, and longitudinal acceleration and its rate of change constraint, which are used to ensure that the trolley remains stable and can drive smoothly during path switching and short-term obstacle avoidance, avoiding safety hazards caused by sudden acceleration or sudden turning of the vehicle.​

[0122] In some embodiments, the system can collect the position, speed, acceleration and path curvature information of the trolley in real time, generate a smooth trajectory according to a prediction model, and dynamically adjust the longitudinal acceleration and the rate of change of curvature, so that the trolley can safely travel along the optimized channel, and can smoothly return to the original channel after short obstacle avoidance.

[0123] In another embodiment, the trajectory curvature constraint can be set to 0.2 to 0.5 , the rate of change of curvature constraint can be set to 0.05 to 0.1 , wherein is meters, is seconds, the longitudinal acceleration constraint can be set to 0.3 to 1.0 m / s², and the acceleration rate constraint can be set to 0.1 to 0.3 m / s³. The trajectory curvature and its rate of change are constrained in the above range based on the comprehensive consideration of the steering ability of the automated guided trolley and the channel width of the working scene, to ensure that the vehicle does not make a sharp turn when switching paths; the longitudinal acceleration and its rate of change are constrained in the above range to maintain smooth acceleration and braking during short obstacle avoidance and return, to avoid impact on the goods, while ensuring safe and efficient operation of the trolley in the channel.

[0124] Preferably, the short-time obstacle avoidance further includes:

[0125] When it is detected that the trolley intersects with a trolley in the opposite direction, the system assigns the trolleys passing priority to perform the vehicle sequential passing operation.

[0126] In an embodiment, when it is detected that the trolley intersects with a trolley in the opposite direction in the spatial channel, the system assigns the trolleys passing priority to perform the vehicle sequential passing operation.

[0127] In some embodiments, the passing priority can be dynamically assigned based on the time sequence of the trolleys entering the intersection, the current speed of the trolleys, and the remaining space in the channel, etc. The trolley with priority passes through the intersection first, and the other trolley waits with a safe distance, and then passes through after the intersection area is safe, to achieve safe and orderly passing. The system can periodically monitor the relative distance and speed of the trolleys in the intersection, to ensure that the priority assignment can be adjusted in real time, to avoid potential collision risks caused by speed differences or delayed judgments of the vehicles.

[0128] In some embodiments, the intersection detection range can be set to 3-6 meters, and the passing priority allocation is triggered when the distance between the vehicles is less than the distance; the safety waiting distance can be set to 0.5-1.0 meters to ensure that the waiting vehicle and the passing vehicle maintain a sufficient safety distance; when the speed difference between the vehicles is more than 0.3 meters per second, the vehicle with higher speed can be allowed to pass the intersection first. The intersection detection range is set to 3-6 meters based on the comprehensive consideration of the braking distance and reaction time of the vehicle, which can detect potential intersection risks in advance; the safety waiting distance is 0.5-1.0 meters, which ensures the safety of the vehicle and does not occupy too much channel space, avoiding affecting the subsequent vehicle passing; the speed difference of 0.3 meters per second is set to reasonably determine the priority passing order when the driving speed of the vehicle fluctuates, and to improve the intersection passing efficiency and safety.

[0129] Especially important is that the detection method of the intersection of the vehicle and the vehicle in the opposite direction further comprises:

[0130] Obtaining the position coordinates of the vehicle and the vehicle in the opposite direction;

[0131] Respectively performing trajectory extension prediction on the position coordinates of the vehicle and the vehicle in the opposite direction to obtain vehicle predicted path data and opposite direction vehicle predicted path data;

[0132] Calculating the path intersection area of the vehicle predicted path data and the opposite direction vehicle predicted path data to generate intersection area candidate data;

[0133] Performing time window overlap detection processing on the intersection area candidate data to obtain the intersection detection result of the vehicle.

[0134] In an embodiment, the trajectory extension prediction can be based on the current speed, acceleration and steering angle of the vehicle to perform multi-step time series prediction, so as to judge the potential risk in advance before the actual intersection occurs; the time window overlap detection is used to judge whether the two vehicles enter the intersection area in the same time period, so as to effectively avoid the misjudgment of only path overlap but time overlap.

[0135] In another embodiment, the time range of trajectory extension prediction can be set to 2-5 seconds to balance the amount of calculation and prediction accuracy; the spatial determination threshold of path intersection area can be set to 0.3-0.8 meters to take into account the vehicle size and positioning accuracy; the time overlap determination threshold of time window overlap detection can be set to 0.5-1.5 seconds to determine whether two vehicles will arrive at the intersection area at the same time. The trajectory prediction time is set to 2-5 seconds because this range can cover the common braking reaction time of the forklift and the medium-short distance intersection scene, and the risk can be predicted in advance. The path intersection area threshold of 0.3-0.8 meters can ensure the safety margin of intersection determination under the premise of considering the actual size of the forklift (about 0.5-0.7 meters) and the positioning error; the time window overlap threshold of 0.5-1.5 seconds can effectively balance the traffic efficiency and safety, and avoid false triggering of obstacle avoidance actions due to extremely short delays.

[0136] Especially important is that the time window overlap detection on the intersection area candidate data further includes:

[0137] According to the intersection area candidate data, the entering time and the leaving time of the forklift and the forklift in the opposite direction in the intersection area are extracted to generate intersection area time interval data;

[0138] The intersection area time interval data is sorted to arrange the entering and leaving records of each forklift in chronological order to generate time sorting list data;

[0139] The time window overlap of adjacent records is detected by using the time sorting list data to generate time window overlap candidate data, to determine whether the stay time intervals of different forklifts overlap;

[0140] According to the time window overlap candidate data, the records with time overlap and the same corresponding intersection area are filtered out to generate effective overlap detection data;

[0141] The effective overlap detection data is converted into forklift intersection event records to generate forklift intersection detection results.

[0142] In an embodiment, according to the intersection area candidate data, the entering time and the leaving time of the forklift and the forklift in the opposite direction in the intersection area are extracted to generate intersection area time interval data. The entering time can be determined by the first intersection time of the vehicle trajectory and the boundary of the intersection area, and the leaving time can be determined by the last intersection time of the vehicle trajectory and the boundary of the intersection area. Secondly, the intersection area time interval data is sorted to arrange the entering and leaving records of each forklift in chronological order to generate time sorting list data. During the sorting process, the time records can be accurate to the millisecond level to improve the detection accuracy.

[0143] Then, time window comparison of adjacent records is performed by using the time-ordered list data to detect whether there is overlap between the stay time intervals of different trolleys, and time window overlap candidate data is generated. The time window comparison can be achieved by judging whether the entering time and leaving time of two trolleys are staggered. Subsequently, records with the same time overlap and corresponding intersection area are filtered out according to the time window overlap candidate data, and valid overlap detection data is generated. This step is used to eliminate misjudgment records with time overlap but located in different intersection areas. Finally, the valid overlap detection data is converted into trolley intersection event records, and trolley intersection detection results are generated. The results can be used as a trigger signal for subsequent passage priority allocation and short-time obstacle avoidance strategy execution.

[0144] In some embodiments, in order to further improve the robustness of detection, the overlap determination of time intervals can set a time overlap determination threshold, for example, 0.5 seconds to 1.5 seconds, and only when the time overlap exceeds the threshold, it is determined as an actual intersection, thereby reducing the false trigger phenomenon caused by small time errors.

[0145] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the application file are intended to be included in the present application.

[0146] The above description is only a specific implementation of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for path planning of an automated guided vehicle based on swarm intelligence, characterized in that, The method comprises the following steps: Step S1: confirming a plurality of candidate paths based on position information between the plurality of automated guided vehicles and the target; confirming a space channel in which the automated guided vehicle can travel according to a center line and a width of the candidate paths; Step S2: acquiring a travel speed of the automated guided vehicle and a congestion degree between the automated guided vehicle and adjacent automated guided vehicles; detecting whether the automated guided vehicle is in a low-speed stable state through the travel speed and the congestion degree; Step S3: screening a stable space channel in the space channel according to the low-speed stable state, and setting a residence time based on the stable space channel; calculating a channel superiority degree of the stable space channel, and confirming whether the automated guided vehicle needs to perform path switching in combination with the residence time; Step S2 comprises the following steps: Step S21: acquiring the travel speed of the automated guided vehicle through a vehicle-mounted motion sensor; Step S22: enabling a vehicle fleet position interconnection function on the automated guided vehicle, and receiving real-time coordinate information periodically broadcast by adjacent automated guided vehicles, wherein the real-time coordinate information comprises a planar position of the automated guided vehicle in a work scene coordinate system; Step S23: for any automated guided vehicle, calculating a straight-line distance between the vehicle and adjacent vehicles according to position information and received position information of the adjacent vehicles, and confirming a relative motion direction of the automated guided vehicle in combination with a change trend of the planar position; Step S24: confirming a space congestion degree of the automated guided vehicle and adjacent vehicles in the same space channel according to the relative motion direction; and detecting whether the automated guided vehicle is in a low-speed stable state according to the travel speed and the congestion degree; Step S24 comprises the following steps: Step S241: judging whether the automated guided vehicle and adjacent vehicles are in the same space channel according to the relative motion direction; Step S242: after determining that the vehicles are in the same space channel, evaluating a space congestion degree of the channel according to a distance between the vehicles, a lateral gap, and a front and rear distribution; Step S243: numerically quantifying the space congestion degree of the channel based on a preset space congestion quantification table to obtain a space congestion degree coefficient; and detecting whether the automated guided vehicle is in a low-speed stable state by using the space congestion degree coefficient and the travel speed.

2. The swarm intelligence based automated guided vehicle path planning method of claim 1, wherein, In Step S243, detecting whether the automated guided vehicle is in a low-speed stable state by using the space congestion degree coefficient and the travel speed comprises: When the travel speed is less than a preset speed threshold value and the space congestion degree coefficient is greater than or equal to a preset congestion threshold value, the automated guided vehicle triggers a low-speed stable mode.

3. The swarm intelligence based automated guided vehicle path planning method of claim 1, wherein, In Step S243, the construction of the space congestion quantification table specifically comprises: collecting vehicle position distribution data in different running scenarios, wherein the vehicle position distribution data comprises a longitudinal distance interval and a lateral gap interval; dividing the space channel into a plurality of levels based on the longitudinal distance interval and the lateral gap interval, and defining a congestion degree identifier for each level; for each congestion degree identifier, performing correlation analysis on a passing speed, a brake frequency, and a waiting time of the vehicle collected in actual operation, extracting a numerical coefficient corresponding to each congestion level, and obtaining the space congestion quantification table.

4. The swarm intelligence based automated guided vehicle path planning method of claim 1, wherein, Step S3 comprises the following steps: Step S31: In the detected low-speed stable state of the trolley, the space channel where the trolley is located is screened out and marked as a candidate stable space channel; Step S32: Vehicle operation anomaly detection is performed on each candidate stable space channel to confirm the stable space channel; and a residence time is allocated to each stable space channel; Step S33: The channel superiority degree of each stable space channel is calculated, and a comprehensive comparison is made in combination with the residence time; when the channel superiority degree is less than a preset superiority degree threshold, and the residence time is greater than or equal to a preset time threshold, a path conversion suggestion signal is generated and sent to the trolley dispatching control module; Step S34: When the channel superiority degree is greater than or equal to the preset superiority degree threshold, or the residence time is less than the preset time threshold, the original channel is kept running.

5. The swarm intelligence based automated guided vehicle path planning method of claim 4, wherein, The vehicle operation anomaly detection on each candidate stable space channel in step S32 includes: Collecting the running data of the trolley in each candidate stable space channel, including the vehicle speed, acceleration, obstacle distance in the channel, and vehicle lateral offset amount; When the vehicle speed of the trolley deviates from the set low-speed interval by more than ±0.2 m / s, or the absolute value of the acceleration continuously exceeds 0.4 m / s² for more than 3 seconds, or the obstacle distance is less than 1.0 m for more than 2 seconds, or the vehicle lateral offset amount exceeds 10 cm, it is determined that the vehicle is running abnormally; In a continuous monitoring period, if no vehicle operation anomaly is detected, the space channel is marked as a stable space channel; if any anomaly is detected in the monitoring period, the space channel is marked as an unstable space channel and is excluded, wherein the continuous monitoring period is 30-120 s.

6. The swarm intelligence based automated guided vehicle path planning method of claim 4, wherein, The formula for calculating the channel superiority degree of each stable space channel in step S33 is as follows: In the formula, is a channel superiority degree coefficient, is a normalized average speed, is a normalized channel congestion coefficient, is a normalized traffic efficiency, is a normalized stability coefficient, is a normalized risk degree, is a weight coefficient, and satisfies .

7. The swarm intelligence based automated guided vehicle path planning method of claim 5, wherein, When the trolley dispatching control module of the trolley receives the path conversion suggestion signal, it further includes: Setting the space channel where the trolley is located as the first channel; Confirming the converted space channel based on the path conversion suggestion signal and setting it as the second channel; Using smooth constraint predictive control for trolley tracking in the second channel and returning to the first channel after completing short-time obstacle avoidance in the second channel, wherein the smooth constraint is specifically the trajectory curvature and its change rate, and the longitudinal acceleration and its change rate constraint.

8. The swarm intelligence based automated guided vehicle path planning method of claim 7, wherein, The short-time obstacle avoidance further includes: When it is detected that the trolley intersects with a trolley in the opposite direction, the trolley and the trolley in the opposite direction are assigned a traffic priority to perform vehicle sequential traffic operation.

Images