AGV driving path control method and device considering energy consumption optimization, equipment and medium

Abstract

The invention discloses an AGV driving path control method and device considering energy consumption optimization, equipment and a medium, and relates to the technical field of AGV control, and the method comprises the following steps: obtaining a first acceleration limit; a first maximum load configuration; performing segmented proportional scaling processing on the first maximum load configuration and the first acceleration limit, and randomly combining to generate a plurality of first energy consumption optimization control sets; aGV driving is controlled according to the first energy consumption optimization control set, feature extraction is carried out on the driving process of each group of AGVs, and an optimization evaluation value is generated through evaluation; and constructing an energy consumption optimization-AGV display model according to each group of first energy consumption optimization control set and the corresponding generated optimization evaluation value, performing data analysis, screening out a second energy consumption optimization control set, and applying the second energy consumption optimization control set to AGV driving path control. The problem of contradiction between energy consumption and stability caused by AGV acceleration and load coupling is effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of AGV control, and more particularly, to an AGV driving path control method and device considering energy consumption optimization, equipment and medium. BACKGROUND

[0002] With the development of intelligent manufacturing and smart logistics, AGV is widely used in warehouse distribution, production line material handling and other scenarios. The existing AGV driving path control method mainly focuses on the shortest path length, the shortest task response time or the optimal obstacle avoidance efficiency, and often uses graph search algorithm, trajectory tracking control, collaborative scheduling optimization and other technologies for path planning and execution control. However, the traditional path control method often ignores the optimization of the overall energy consumption of the system, especially in the application environment of frequent start-stop of multi-AGV system, significant load change and complex path, the actual energy consumption of AGV is closely related to path selection and control parameter setting.

[0003] In energy consumption optimization control, some existing technologies begin to introduce speed limit, obstacle avoidance frequency control and other strategies to reduce unnecessary energy consumption, but there is still a lack of systematic coupling modeling and regulation method for two key parameters with actual physical control significance, namely acceleration limit and maximum load configuration, during the AGV running process. These two parameters can be directly set by the control system, and they will also significantly affect the energy consumption of AGV in the aspects of acceleration stage, turning path selection, slope passing and start-stop frequency. The existing path control strategy fails to fully consider the collaborative constraints of acceleration and load at the path execution level, which easily leads to unstable energy consumption control and decreased running efficiency.

[0004] Specifically, acceleration limit and maximum load configuration have obvious defects when set at high and low values. If the acceleration limit is set too high, the motor power demand of AGV increases during startup and acceleration, and the energy consumption increases significantly, while there is a risk of slipping, losing control or tire wear too fast; if it is set too low, it will cause slow start, affect the task response speed, and even cause power shortage in complex paths such as climbing. Similarly, if the AGV load configuration is too large, although the single task efficiency is higher, the overall load increase will cause the energy consumption to increase sharply, and aggravate the problems of climbing difficulty, turning instability and other problems; on the contrary, if the load configuration is too low, the transportation frequency needs to be increased, which leads to overall path scheduling redundancy, and the proportion of empty running increases, and the unit task energy consumption actually increases.

[0005] There is a significant mutual coupling between the speed limit and the maximum load configuration. If the acceleration is not properly adjusted under high load conditions, the AGV may be unstable, energy waste and hardware loss during high inertia start-up. If the high load is maintained under low acceleration conditions, the task may fail, the path may be invalid or the climbing may be limited due to insufficient power. Conversely, under light load conditions, increasing acceleration can improve response efficiency, but may also cause energy conversion efficiency to decrease and path tracking accuracy to decrease. Therefore, the two physical control parameters need to be coordinated to achieve energy efficiency balance in path control.

[0006] For example: for warehouse AGV cross-floor transportation tasks, the task weight is large, the path needs to pass through a slope, the load is large and the acceleration must be low, otherwise the vehicle will overload and lose control when climbing the slope. For e-commerce sorting light load multi-frequency tasks, the light load is frequent start and stop, the acceleration can be adjusted to be high, but the maximum acceleration time period still needs to be limited for energy saving, the load is small, but the start and stop are frequent, and the acceleration frequency needs to be balanced to control energy consumption.

[0007] In the prior art, the control of AGV acceleration limit and maximum load configuration is mostly independent setting, lacking systematic analysis and joint optimization of the coupling relationship between the two, which can easily lead to insufficient power, high energy consumption or path failure during operation, and it is difficult to achieve stable path control with optimal energy consumption.

[0008] To solve the above problems, the present application provides a solution. SUMMARY

[0009] To overcome the above-mentioned defects of the prior art, embodiments of the present application provide an AGV driving path control method, device, equipment and medium considering energy consumption optimization, which effectively solves the contradiction between energy consumption and stability caused by the coupling of AGV acceleration and load through multi-dimensional feature evaluation and optimization control set screening.

[0010] To achieve the above-mentioned purpose, the present application provides the following technical solutions: The application discloses an AGV driving path control method considering energy consumption optimization, which comprises the following steps: obtaining a first path, initializing an acceleration limit based on the curve characteristics of the first path to obtain a first acceleration limit; estimating a maximum load according to the terrain slope and key node load demand of the first path to obtain a first maximum load configuration; defining an energy consumption optimization control set, wherein the energy consumption optimization control set comprises a maximum load configuration and an acceleration limit; performing segmented proportional scaling processing on the first maximum load configuration and the first acceleration limit, and randomly combining to generate a plurality of first energy consumption optimization control sets; controlling AGV driving according to the first energy consumption optimization control set, extracting features of each AGV driving process, and evaluating an optimization evaluation value; constructing an energy consumption optimization-AGV display model according to each first energy consumption optimization control set and the corresponding generated optimization evaluation value, and performing data analysis to screen out a second energy consumption optimization control set, which is applied to AGV driving path control.

[0011] In a preferred embodiment, the first path is a pre-acquired initial AGV path; the spatial coordinate point sequence of the first path is acquired; the corner change between adjacent coordinate points on the path is calculated; the turning radius of each point on the path is calculated based on the corner change; the minimum turning radius in the path is extracted; the maximum lateral acceleration range allowed is determined according to the minimum turning radius and in combination with the lateral stability requirement of the vehicle; the maximum acceleration limit is set in combination with the longitudinal dynamic performance of the vehicle, so that the maximum acceleration limit does not exceed the maximum lateral acceleration range; and the maximum acceleration limit is taken as the first acceleration limit.

[0012] In a preferred embodiment, the maximum load is estimated according to the terrain slope and key node load demand of the first path to obtain a first maximum load configuration, specifically: the slope data of the first path is acquired, including the slope values of each position point on the path; the maximum continuous interval of the slope in the first path is identified, and the maximum slope value of the interval is extracted; the maximum load that can be carried by the vehicle under the condition of the maximum slope for safe climbing is calculated as the maximum load value under the slope limit in combination with the driving force model and dynamic parameters of the vehicle; the design maximum carrying mass of all bridges in the first path is acquired, and the minimum bridge carrying mass value is extracted; the maximum load limit of all narrow passages in the path is acquired, and the minimum narrow passage load limit value is extracted; the minimum value is selected as the maximum load limit value of the key node of the path from the maximum load limit of the slope, the minimum bridge carrying mass and the minimum narrow passage load limit.

[0013] In a preferred embodiment, the first maximum load configuration and the first acceleration limit are segmented and scaled proportionally, and a plurality of first energy consumption optimization control sets are generated by random combination, specifically: a scaling sequence of the acceleration limit is set, the scaling sequence includes a plurality of scaling factors, the scaling factors are based on the first acceleration limit, and a group of acceleration limit candidate values are generated by scaling; a scaling sequence of the maximum load configuration is set, the scaling sequence includes a plurality of scaling factors, the scaling factors are based on the first maximum load configuration, and a group of maximum load configuration candidate values are generated by scaling; based on the acceleration limit candidate values and the maximum load configuration candidate values, a plurality of combinations of the acceleration limit and the maximum load configuration are generated by random combination as the first energy consumption optimization control set.

[0014] In a preferred embodiment, the features include energy consumption features, turning stability features, and task response speed features; the specific acquisition method of the energy consumption features is as follows: real-time current and voltage data of the driving motor of the AGV during the execution of the path task are collected; based on the collected current and voltage data, the instantaneous electric power at each sampling time is calculated; the instantaneous power in the entire task execution time interval is integrated to obtain the total energy consumption during the task; the cumulative travel distance of the AGV during the task execution is recorded; the unit distance energy consumption is calculated, and the total energy consumption is divided by the cumulative travel distance to eliminate the influence of the path length difference; the load weight change data during the task is obtained, and the load weighting coefficient is calculated, which is the time average value of the ratio of the load weight during the task to the maximum load; the unit distance energy consumption and the load weighting coefficient are combined, and weighted calculation is performed according to the set weight coefficient to obtain the energy consumption features.

[0015] In a preferred embodiment, the specific acquisition method of the task response speed features is as follows: response time data from receiving a scheduling instruction to actually starting an action during the execution of the task by the AGV is collected; the completion time of each key node during the task execution is recorded; the average value of all response times and key node completion times is calculated; the standard deviation of the response time and the key node completion time is calculated to reflect the fluctuation of the response time; the response speed variation coefficient is calculated by using the average value and the standard deviation as the task response speed features.

[0016] In a preferred embodiment, the energy consumption optimization-AGV display model is constructed according to each set of first energy consumption optimization control set and corresponding generated optimization evaluation value, specifically: each set of energy consumption optimization control set and its corresponding optimization evaluation value are one-to-one corresponding, forming a mapping relationship of energy consumption optimization control set and optimization evaluation value, that is, a point;The energy consumption optimization-AGV display model is a multi-dimensional parameter space, which is used to intuitively present the distribution of different control sets in energy consumption and performance;The mapping relationship of several energy consumption optimization control sets and optimization evaluation values is mapped to the energy consumption optimization-AGV display model, and several points are fitted, and the energy consumption optimization control set corresponding to the peak value is taken as the second energy consumption optimization control set and applied to the AGV driving path control.

[0017] An AGV driving path control system considering energy consumption optimization, comprising an acceleration limit initialization module, a maximum load configuration initialization module, an energy consumption optimization control set acquisition module, a feature extraction module and a screening control module;The acceleration limit initialization module is used to obtain a first path, initialize the acceleration limit based on the curve characteristics of the first path, and obtain a first acceleration limit;The maximum load configuration initialization module is used to estimate the maximum load according to the terrain slope and key node load demand of the first path, and obtain a first maximum load configuration;The energy consumption optimization control set acquisition module is used to define an energy consumption optimization control set, which comprises a maximum load configuration and an acceleration limit;The first maximum load configuration and the first acceleration limit are subjected to segmented proportional scaling processing and randomly combined to generate several first energy consumption optimization control sets;The feature extraction module is used to control the AGV driving according to the first energy consumption optimization control set, and to extract features from the AGV driving process of each group and generate optimization evaluation values;The screening control module is used to construct an energy consumption optimization-AGV display model according to each set of first energy consumption optimization control set and corresponding generated optimization evaluation value, and to analyze the data and screen out a second energy consumption optimization control set for AGV driving path control.

[0018] An electronic device, comprising: at least one processor;And a memory connected in communication with the at least one processor;Wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the AGV driving path control method considering energy consumption optimization.

[0019] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the AGV driving path control method considering energy consumption optimization.

[0020] The technical effects and advantages of the AGV driving path control method, device, equipment and medium considering energy consumption optimization of the application are as follows: The application scientifically sets the acceleration limit and maximum load configuration of the AGV by systematically combining the path curve characteristics, terrain slope and key node load limit, and then constructs multiple energy consumption optimization control sets, and comprehensively evaluates and screens based on multi-dimensional characteristics such as energy consumption, turning stability and task response speed, so as to ensure that the optimized control parameters realize the optimal comprehensive performance of the lowest energy consumption, high driving stability and fast task response under the premise of guaranteeing the vehicle dynamics safety and path physical constraints. The method effectively solves the complex coupling problem between acceleration and load, improves the safety and efficiency of AGV path execution, enhances the adaptability and robustness of the control strategy, and promotes the efficient and stable operation of intelligent logistics vehicles. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 A flowchart of an AGV driving path control method considering energy consumption optimization.

[0022] Figure 2 A structural schematic diagram of an AGV driving path control device considering energy consumption optimization.

[0023] Figure 3 An electronic device structural schematic diagram of an AGV driving path control method considering energy consumption optimization. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0025] Embodiment 1, Figure 1 The AGV driving path control method considering energy consumption optimization is given, including the following steps: S1, obtaining a first path, initializing an acceleration limit based on the curve characteristics of the first path to obtain a first acceleration limit.

[0026] In this embodiment, the first path is an initial AGV path obtained in advance, the space coordinate point sequence of the first path is obtained, the turning angle change between adjacent coordinate points on the path is calculated, the turning radius of each point on the path is calculated based on the turning angle change, and the first acceleration limit is obtained based on the curve characteristics of the first path. The first path is a pre-obtained initial AGV path; The space coordinate point sequence of the first path is obtained; The turning angle change between adjacent coordinate points on the path is calculated; The turning radius of each point on the path is calculated based on the turning angle change; The minimum turning radius in the extraction path; According to the minimum turning radius, in combination with the vehicle lateral stability requirement, the maximum lateral acceleration range allowed is determined; In combination with the vehicle longitudinal dynamic performance, the maximum acceleration limit is set so as not to exceed the maximum lateral acceleration range; The maximum acceleration limit is taken as the first acceleration limit.

[0027] It should be noted that the determination of the maximum lateral acceleration range can be based on the calculation of the tire grip of the vehicle and the vehicle speed, specifically that the lateral centrifugal force generated by the vehicle during turning cannot exceed the maximum friction between the tire and the ground, so as to ensure the turning stability of the vehicle. The maximum lateral acceleration can be obtained through empirical data, or determined based on a vehicle dynamics model, in combination with parameters such as tire friction coefficient, vehicle gravity center height, etc.

[0028] It should be noted that the setting of the maximum acceleration limit not only considers the maximum lateral acceleration range, but also combines factors such as the maximum output capability of the vehicle motor and driving system, energy consumption, and mechanical safety limit, to ensure that the acceleration limit does not exceed the safety boundary of the vehicle longitudinal dynamic performance. The limit value can be obtained through test data, or can be dynamically adjusted in actual operation. The following is a feasible first acceleration limit acquisition formula: Maximum lateral acceleration range: ; Actual lateral acceleration of the vehicle during turning: ; Maximum longitudinal acceleration allowed: ; In combination with the maximum output capability of the motor, the smaller value is finally taken as the first acceleration limit: ; In the formula, is the friction coefficient between the tire of the vehicle and the ground; g is the acceleration of gravity; v is the maximum estimated driving speed of the vehicle on the first path; is the minimum turning radius on the first path.

[0029] It should be noted that the turning radius of the path is approximately obtained by calculating the circular arc radius formed by three consecutive points on the path, and the turning angle change is the measurement of the included angle of adjacent path segments. This method is a commonly used path curve characteristic analysis technique. The first path is a discrete set of spatial coordinate points, which can be obtained through sensor acquisition or planning system output.

[0030] S2, according to the terrain slope of the first path and the load demand of the key nodes, estimate the maximum load to obtain the first maximum load configuration; In the embodiment, the maximum load is estimated according to the terrain slope of the first path and the load demand of the key node, and a first maximum load configuration is obtained, specifically: Obtain the slope data of the first path, including the slope values of each position point on the path; Identify the continuous interval with the maximum slope in the first path, and extract the maximum slope value of the interval; Combine the driving force model and dynamic parameters of the vehicle to calculate the maximum load that the vehicle can carry under the condition of the maximum slope, as the maximum load value under the slope restriction; Obtain the design maximum load capacity of all bridges in the first path, and extract the minimum bridge load capacity value; Obtain the maximum load limit of all narrow passages in the path, and extract the minimum narrow passage load limit value; From the maximum load limit under the slope restriction, the minimum bridge load capacity and the minimum narrow passage load limit, select the minimum value as the maximum load limit value of the key node of the path; Compare the rated maximum load designed for the vehicle with the maximum load limit value of the key node of the path, and take the smaller value as the first maximum load configuration. The following is the formula of the feasible first maximum load configuration: , wherein is the maximum load capacity calculated by the vehicle according to the driving force model in the maximum slope interval of the first path; is the minimum design load capacity of all bridges in the first path, is the minimum allowable load limit value of all narrow passages in the first path, is the rated maximum load value specified in the vehicle design parameters, and the calculation formula of the feasible maximum load capacity is as follows (without considering air resistance): , wherein is the maximum driving force, is the slope angle, is the rolling resistance coefficient (generally 0.01~0.03), is the vehicle weight.

[0031] It should be noted that the slope data can be obtained based on the first path through the path design system, or measured by using topographic maps in combination with laser radar, high-precision GPS and other equipment. The slope value of each position point on the path can be represented as the ratio of the height difference within a certain distance before and after the point to the horizontal distance. The obtained continuous slope data is used for subsequent slope interval identification and analysis.

[0032] It should be noted that the identification of the maximum slope interval refers to filtering all continuous uphill sections from the path, calculating the average slope and maximum slope value for each section, and then extracting the overall maximum slope value as the boundary condition for subsequent load capacity analysis, to ensure that the vehicle can still operate normally under the most adverse slope.

[0033] It should be noted that the safe load of the vehicle under the maximum slope can be combined with the maximum output torque of the vehicle drive motor, the friction coefficient between the tire and the ground, the total mass of the vehicle, and the wheel diameter, etc. parameters, to estimate the maximum effective traction force that can be achieved on the slope through the vehicle driving capability model, and to convert the allowed maximum total load mass, to ensure that the vehicle does not slip or power shortage under the slope.

[0034] It should be noted that the maximum load capacity of the bridge is a structural design parameter provided by the path planning system or the map database in advance, which is usually calibrated or recorded by the infrastructure maintenance unit. If the path contains multiple bridges, the lowest load capacity among them is extracted as the maximum allowed load under the bridge restriction condition.

[0035] It should be noted that the maximum load limit of the narrow passage is determined according to the maximum bearing capacity of the structure on both sides of the passage or the requirement for safe passage, which is provided by the factory transportation specification or engineering design data. If there are multiple narrow passages in the path, the minimum load limit among them is also extracted as the constraint condition.

[0036] It should be noted that the maximum load limit value of the path key node is the most stringent condition among the three types of constraint factors mentioned above, that is, the minimum value among the three is taken as the maximum running load limit that the path can withstand. This method can ensure that the vehicle safely passes through all physical structures and dynamic conditions without structural overload or passage failure.

[0037] It should be noted that the rated maximum load of the vehicle is determined by the vehicle design parameters, which are usually provided by the manufacturer. If the value is higher than the path key node limit value, it must be adjusted to adapt to the path characteristics, and the smaller one of the two is finally selected as the first maximum load configuration to ensure that both path safety and vehicle stability requirements are met.

[0038] It should be noted that the first path is the initial default path, which is generated by the AGV based on the preset target area and job requirements before the task starts or at the initial stage of scheduling planning. This path can be generated from the map data before the task is issued through the path planning algorithm, including the main path coordinates of the area to be traveled, structure information and known constraint conditions. In the actual running process, this path can be adjusted in real time according to dynamic obstacle avoidance, task update, traffic scheduling changes, etc. Therefore, the first path does not necessarily cover all potential path characteristics.

[0039] And in the estimation of the maximum load configuration, if the first path does not contain part of the third type of constraint factors (such as missing bridge information or narrow channel restriction information), the default is that this factor does not constitute a constraint, that is, the limit is set to infinity or a value greater than the maximum design load of the vehicle, ensuring that the lack of this information does not affect the evaluation of the maximum load configuration by other factors, and the overall estimation result is still valid and executable.

[0040] S3, define an energy consumption optimization control set, which includes a maximum load configuration and an acceleration limit; S4, segment and scale the first maximum load configuration and the first acceleration limit, and randomly combine to generate a plurality of first energy consumption optimization control sets; In this embodiment, the first maximum load configuration and the first acceleration limit are segmented and scaled, and a plurality of first energy consumption optimization control sets are randomly combined, specifically: Set the scaling ratio sequence of the acceleration limit, which includes a plurality of ratio factors, and the ratio factors generate a group of acceleration limit candidate values based on the first acceleration limit; Set the scaling ratio sequence of the maximum load configuration, which includes a plurality of ratio factors, and the ratio factors generate a group of maximum load configuration candidate values based on the first maximum load configuration; Based on the acceleration limit candidate value and the maximum load configuration candidate value, a plurality of combinations of acceleration limit and maximum load configuration are generated by random combination as the first energy consumption optimization control set.

[0041] It should be noted that the ratio factors in the scaling ratio sequence of the acceleration limit are selected in the range of 0.7 to 1.3, which are set by equal interval or experience value, to ensure that the generated candidate acceleration limit can cover the low-speed stable interval required for safe operation of the vehicle, and also include higher acceleration to meet the dynamic response demand. It should be noted that the ratio factors in the scaling ratio sequence of the maximum load configuration are selected in the range of 0.6 to 1.2, and the specific values of the ratio factors are adjusted according to the vehicle design rated load and the maximum load limit value of the path key node.

[0042] It should be noted that the number of combinations of acceleration limit and maximum load configuration generated by random combination is determined according to actual optimization demand and calculation resources, which is not less than 10 groups and not more than 50 groups, to balance the coverage of the search space and the calculation efficiency.

[0043] It should be noted that each combination generated in the first energy consumption optimization control set needs to meet the vehicle dynamics safety constraints and path physical constraints to ensure that the acceleration limit and load configuration corresponding to the combination are feasible and safe in actual operation.

[0044] S5, controlling the AGV to travel according to the first energy consumption optimization control set, and performing feature extraction on each AGV travel process to generate an optimization evaluation value; In the embodiment, the feature extraction on each AGV travel process is specifically: The features include energy consumption features, turning stability features, and task response speed features.

[0045] The energy consumption feature is an index value for quantifying the overall energy consumption level of the AGV in performing the path task under different acceleration limit and maximum load configuration combinations. This feature is obtained by collecting real-time current, voltage, driving power or battery power change parameters of the AGV in the entire travel path, combining with the vehicle running time and path length, calculating the total electric energy consumed per task or the energy consumption per unit distance, and finally obtaining a numerical feature for evaluating energy efficiency performance.

[0046] Analyzing the energy consumption feature has the following advantages for evaluating the effects of different energy consumption optimization control sets and for screening and finally solving the coupling problem of acceleration and load: Quantitative comparison of control sets: through the extraction and analysis of the energy consumption feature index, the energy consumption performance of different control strategies under the same working condition can be quantitatively evaluated, thereby providing a unified criterion for screening the optimal control set.

[0047] Revealing the nonlinear coupling effect of acceleration and load on energy consumption: energy consumption feature analysis can reveal the common influence mode of acceleration change (such as start frequency, acceleration and deceleration gradient) and load fluctuation (such as load step change) on energy consumption under the control set, which helps to identify the sensitive area of the coupling of the two on system energy efficiency.

[0048] Supporting the generation of adaptive control factors: through the mapping relationship between the energy consumption feature and the control parameters, a regression model is established, which can be used to guide the design of the adjustment factor of the subsequent optimization control set, and realize the energy efficiency optimal response for different load-acceleration combinations.

[0049] Enhancing the robustness and generalization ability of the control strategy: analyzing and extracting stable energy consumption features under various working conditions makes the selected control set more robust and can adapt to typical and abnormal load-acceleration coupling situations, avoiding overfitting optimization under a single working condition.

[0050] Promote multi-objective collaborative optimization: Incorporate energy consumption characteristics into the control set optimization evaluation system, which can realize the trade-off between energy consumption, response stability, mechanical impact and other multiple objectives, and help to optimize the overall performance of the final control strategy.

[0051] The specific acquisition method of the energy consumption characteristic is as follows: Collect the real-time current and voltage data of the driving motor when the AGV executes the path task; Based on the collected current and voltage data, the instantaneous electric power at each sampling time is calculated; Integrate the instantaneous power in the entire task execution time interval to obtain the total energy consumption during the task; Record the cumulative travel distance of the AGV during the task execution; Calculate the energy consumption per unit distance, divide the total energy consumption by the cumulative travel distance to eliminate the influence of path length difference; Obtain the load weight change data during the task, calculate the load weighting coefficient, which is the time average of the ratio of the load weight during the task to the maximum load; Combine the energy consumption per unit distance and the load weighting coefficient, and calculate the energy consumption characteristic according to the set weight coefficient.

[0052] The specific calculation formula of the energy consumption characteristic is as follows: Instantaneous electric power: ; Total energy consumption: ; Energy consumption per unit distance: ; Load weighting coefficient: ; Energy consumption characteristic: ; In the formula, is the voltage; is the current; is the task time; is the load weight during the task; is the maximum load configuration; is the load influence weight coefficient, the value range is 0.1 to 0.5, the default is 0.3, and D is the cumulative travel distance.

[0053] It should be noted that the calculation of total energy consumption uses the integral method to accumulate the instantaneous electric power in the time domain, and the integral interval should strictly correspond to the time range from the start of the task to the end of the task. The time window is usually provided by the task scheduling system or the upper control system, which ensures that the energy consumption calculation is strictly aligned with the path task, avoids including the energy consumption of idle or standby phase, and affects the evaluation accuracy.

[0054] It should be noted that the cumulative distance of the AGV travel path is calculated in real time by the internal odometer or inertial navigation system, and the error should be within 2%. If the system has a large ranging error, a mileage correction module should be introduced to ensure the credibility of the unit distance energy consumption.

[0055] It should be noted that the setting of the load influence weight coefficient is determined by engineering experience data. A higher weight indicates a higher sensitivity to load changes on energy consumption, which is suitable for logistics task scenarios with large load fluctuations. A lower weight is suitable for application environments with stable load and high path complexity.

[0056] The turning stability feature is an index value for quantifying the ability of AGV to maintain posture balance and tire adhesion force control under different acceleration limit and maximum load configuration combinations during path turning. This feature collects dynamic parameters such as yaw rate, lateral acceleration, and steering trajectory offset of AGV in the path turning section, evaluates whether the AGV has side slip, deviation, or lateral instability during turning, and extracts numerical features of stability performance by combining the maximum roll angle or attitude fluctuation amplitude of the vehicle when passing through a typical bend. This feature can reflect the direct impact of acceleration limit and load configuration on vehicle turning safety.

[0057] Analyzing the turning stability feature to evaluate the effectiveness of different energy consumption optimization control sets has the following advantages for screening and ultimately solving the coupling problem of acceleration and load: Expanded evaluation dimensions: not only limited to energy consumption, but also covering dynamic stability, safety, etc.

[0058] More accurate control set screening: through the response analysis of the vehicle during turning, it can be found that some control strategies are energy-saving but have poor stability, helping to eliminate unsuitable solutions.

[0059] Coupling problem visualization: the coupling of acceleration and load is essentially reflected in actual maneuverability, and this coupling can be quantitatively reflected through the stability index.

[0060] Promote systematic optimization: not only optimize energy consumption, but also ensure the dynamic performance of the whole vehicle, achieving multi-objective coordinated control.

[0061] The specific acquisition method of the turning stability feature is as follows: Collect the lateral acceleration data and steering angle data of each turning section during the AGV executing the path task; Identify all turning sections in the path and extract the maximum value of lateral acceleration in each turning section; Compare the maximum value of lateral acceleration of each turning section with the vehicle lateral stability threshold to determine whether it exceeds the threshold; Count the number of times the lateral acceleration exceeds the stability threshold in all turning sections as the number of abnormal turning stability times; Record the average value of the actual driving speed in all turning sections and compare it with the upper limit of the safe turning speed allowed by the vehicle design; Count the number of turning sections whose speed exceeds the upper limit of the safe speed as the number of speeding turning times; Add the number of abnormal turning stability times and the number of speeding turning times to get the total number of unstable turning events; Calculate the unstable turning event ratio as the turning stability characteristic value by dividing the number of unstable turning events by the total number of turning sections.

[0062] It should be noted that the lateral acceleration data refers to the acceleration value perpendicular to the driving direction of the vehicle detected by the inertia sensor (such as an accelerometer) when the vehicle is turning, which is used to reflect the lateral force of the vehicle during turning and evaluate its lateral stability.

[0063] It should be noted that the vehicle lateral stability threshold is a lateral acceleration limit value preset according to factors such as vehicle chassis design, tire grip, center of gravity height, and ground adhesion coefficient. Exceeding this value may cause the vehicle to slide, deflect, or lose control. This threshold can be obtained through simulation and field testing.

[0064] It should be noted that the actual driving speed is the instantaneous speed value collected by the AGV body speed sensor in real time. The time average of this speed data in each turning section can obtain the average turning speed of the section to reflect the dynamic performance of the AGV during turning.

[0065] It should be noted that the upper limit of the safe turning speed is a theoretical value calculated according to the vehicle dynamics model, turning radius, and maximum allowable lateral acceleration, which is used to determine whether the AGV has a speed too fast behavior during turning. It is preset by the vehicle control system.

[0066] It should be noted that the unstable turning event ratio as the turning stability characteristic value has the normalization characteristic, which can be applied to different lengths and different structure complexity path tasks, so that the turning stability results of different tasks have comparability.

[0067] The task response speed feature is used to quantify the response efficiency of the AGV after receiving the scheduling instruction to complete the corresponding action. This feature measures the delay time from the instruction issuance to the actual start of the AGV action, as well as the completion time of each key node in the task execution process, to comprehensively reflect the response speed performance of the AGV under different acceleration limits and maximum load configuration conditions.

[0068] The task response speed characteristic has the following advantages for evaluating the effects of different energy consumption optimization control sets, screening, and finally solving the coupling problem of acceleration and load: Directly reflects the dynamic response capability of AGV under different control parameter combinations, helping to identify key factors affecting operation efficiency.

[0069] Reveals the impact of acceleration limit on AGV start-up and acceleration process, avoiding task delays caused by slow response.

[0070] Reflects the impact of maximum load configuration on mechanical burden and power system response speed, helping to balance load and acceleration coordination.

[0071] Promotes effective control of task execution timeliness while ensuring energy consumption optimization.

[0072] Helps to screen parameter combinations that can reduce energy consumption while maintaining fast response, improving overall operation efficiency.

[0073] The specific method for obtaining the task response speed characteristic is as follows: Collect the response time data from receiving the dispatching instruction to actually starting the action during the AGV executing the task; Record the completion time of each key node during the task execution process; Calculate the average of all response times and key node completion times; Calculate the standard deviation of response time and key node completion time to reflect the fluctuation of response time; Use the average and standard deviation to calculate the response speed coefficient of variation as the task response speed characteristic.

[0074] The evaluation generates an optimization evaluation value, specifically: Based on the extracted energy consumption characteristic, turning stability characteristic, and task response speed characteristic, a standard optimization evaluation model is constructed; The standard optimization evaluation model is trained based on machine learning, adjusting the model parameters to fit the relationship between each characteristic and the optimization target; The best weight parameters of the standard optimization evaluation model are solved through the training process; The solved optimization evaluation model takes the maximum acceleration limit and maximum load configuration in the control set as input and outputs the corresponding optimization evaluation value; This optimization evaluation value is used to quantify the comprehensive performance of AGV driving under different control sets, guiding subsequent parameter selection and optimization; The standard optimization evaluation model is a weighted form of the energy consumption characteristic, turning stability characteristic, and task response speed characteristic, and the target of machine learning training is the corresponding best weight.

[0075] S6, according to each set of first energy consumption optimization control set and the corresponding generated optimization evaluation value, construct energy consumption optimization-AGV display model, and carry out data analysis, screen out second energy consumption optimization control set, and apply to AGV driving path control; In the embodiment, the energy consumption optimization-AGV display model is constructed according to each set of first energy consumption optimization control set and the corresponding generated optimization evaluation value, specifically: Each set of energy consumption optimization control set and its corresponding optimization evaluation value are one-to-one corresponding, forming a mapping relationship between energy consumption optimization control set and optimization evaluation value, that is, a point; The energy consumption optimization-AGV display model is a multi-dimensional parameter space, which is used to intuitively present the distribution of different control sets in energy consumption and performance; The mapping relationship between the plurality of energy consumption optimization control sets and the optimization evaluation values is mapped to the energy consumption optimization-AGV display model, and the plurality of points are fitted, and the energy consumption optimization control set corresponding to the peak value is taken as the second energy consumption optimization control set and applied to the AGV driving path control.

[0076] The present application scientifically sets the acceleration limit and maximum load configuration of AGV by systematically combining the path curve characteristics, terrain slope and key node load limit, and then constructs a plurality of energy consumption optimization control sets, and comprehensively evaluates and screens based on energy consumption, turning stability and task response speed and other multi-dimensional characteristics, to ensure that the optimized control parameters can ensure the vehicle dynamics safety and path physical constraints, and realize the optimal comprehensive performance of the lowest energy consumption, high driving stability and fast task response. The method effectively solves the complex coupling problem between acceleration and load, improves the safety and efficiency of AGV path execution, enhances the adaptability and robustness of the control strategy, and promotes the efficient and stable operation of intelligent logistics vehicles.

[0077] Embodiment 2, Figure 2 An AGV driving path control system considering energy consumption optimization is given, which includes an acceleration limit initialization module, a maximum load configuration initialization module, an energy consumption optimization control set acquisition module, a feature extraction module and a screening control module. The acceleration limit initialization module is used for obtaining a first path, initializing the acceleration limit based on the curve characteristics of the first path, and obtaining a first acceleration limit; The maximum load configuration initialization module is used for estimating the maximum load according to the terrain slope and key node load demand of the first path, and obtaining a first maximum load configuration; The energy consumption optimization control set acquisition module is used for defining an energy consumption optimization control set, the energy consumption optimization control set including a maximum load configuration and an acceleration limit;The first maximum load configuration and the first acceleration limit are subjected to segmented proportional scaling processing, and a plurality of first energy consumption optimization control sets are randomly combined. a feature extraction module, configured to control the AGV to travel according to the first energy consumption optimization control set, and to extract features from each set of AGV travel process, and to evaluate and generate an optimization evaluation value; a screening control module, configured to construct an energy consumption optimization-AGV display model according to each set of the first energy consumption optimization control set and the corresponding generated optimization evaluation value, and to perform data analysis, and to screen out a second energy consumption optimization control set, and to apply the second energy consumption optimization control set to AGV travel path control.

[0078] The application also includes an electronic device, characterized in that the electronic device comprises: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the AGV travel path control method considering energy consumption optimization.

[0079] The application also includes a computer readable storage medium storing a computer program, characterized in that the computer program is executed by a processor to implement the AGV travel path control method considering energy consumption optimization.

Claims

1. A method for controlling a travel path of an AGV with consideration of energy consumption optimization, characterized by, The method comprises the following steps: obtaining a first path, initializing an acceleration limit based on the curve characteristics of the first path, and obtaining a first acceleration limit; estimating the maximum load based on the terrain slope and key node load demand of the first path, and obtaining a first maximum load configuration; defining an energy consumption optimization control set, which includes a maximum load configuration and an acceleration limit; segmenting and scaling the first maximum load configuration and the first acceleration limit, and randomly combining to generate a plurality of first energy consumption optimization control sets; controlling the AGV to travel according to the first energy consumption optimization control set, and extracting features from the AGV travel process to evaluate the optimization evaluation value; constructing an energy consumption optimization-AGV display model according to each first energy consumption optimization control set and the corresponding generated optimization evaluation value, and performing data analysis to screen out a second energy consumption optimization control set for AGV travel path control.

2. The AGV travel path control method considering energy consumption optimization according to claim 1, characterized by, The method comprises the following steps: The first path is a pre-obtained initial AGV path; obtaining a sequence of spatial coordinate points of the first path; calculating the change of the angle between adjacent coordinate points on the path; calculating the turning radius of each point on the path based on the change of the angle; extracting the minimum turning radius in the path; determining the maximum lateral acceleration range allowed according to the minimum turning radius and the vehicle lateral stability requirement; combining the vehicle longitudinal dynamic performance to set the maximum acceleration limit so that it does not exceed the maximum lateral acceleration range; the maximum acceleration limit is used as the first acceleration limit.

3. The AGV travel path control method considering energy consumption optimization according to claim 2, characterized by, The method comprises the following steps: obtaining the slope data of the first path, including the slope values of each position point on the path; identifying the maximum continuous interval of the slope in the first path and extracting the maximum slope value of the interval; combining the vehicle drive force model and dynamic parameters to calculate the maximum load that the vehicle can carry safely under the condition of the maximum slope, as the maximum load value under the slope limit; obtaining the design maximum carrying capacity of all bridges in the first path, and extracting the minimum bridge carrying capacity value; obtaining the maximum load limit of all narrow passages in the path, and extracting the minimum narrow passage load limit value; selecting the minimum value from the maximum load limit of the slope, the minimum bridge carrying capacity and the minimum narrow passage load limit as the maximum load limit value of the key node of the path.

4. The AGV travel path control method considering energy consumption optimization according to claim 3, characterized by, The method comprises the following steps: setting the scaling ratio sequence of the acceleration limit, which includes a plurality of scaling factors, and generating a group of acceleration limit candidate values based on the first acceleration limit; setting the scaling ratio sequence of the maximum load configuration, which includes a plurality of scaling factors, and generating a group of maximum load configuration candidate values based on the first maximum load configuration; Based on the acceleration limit candidate value and the maximum load configuration candidate value, a plurality of combinations of acceleration limit and maximum load configuration are generated by random combination as a first energy consumption optimization control set.

5. The AGV travel path control method considering energy consumption optimization according to claim 4, characterized by, The features include energy consumption features, turning stability features, and task response speed features. The specific acquisition method of the energy consumption feature is as follows: Real-time current and voltage data of the driving motor when the AGV executes the path task are collected. Based on the collected current and voltage data, the instantaneous electric power at each sampling time is calculated. The total energy consumption during the task is obtained by integrating the instantaneous power over the entire task execution time interval. The cumulative travel distance of the AGV during the task execution is recorded. The unit distance energy consumption is calculated by dividing the total energy consumption by the cumulative travel distance to eliminate the influence of path length differences. Load weight change data during the task is obtained, and a load weighting coefficient is calculated, which is the time average of the ratio of the load weight to the maximum load during the task. The unit distance energy consumption and the load weighting coefficient are combined and weighted according to the set weight coefficient to obtain the energy consumption feature.

6. The AGV travel path control method considering energy consumption optimization according to claim 5, characterized by, The specific acquisition method of the task response speed feature is as follows: Response time data from receiving the dispatching instruction to actually starting the action during the AGV executing the task are collected. The completion times of each key node during the task execution are recorded. The average of all response times and key node completion times is calculated. The standard deviation of the response time and the key node completion time is calculated to reflect the fluctuation of the response time. The response speed variation coefficient is calculated using the average and the standard deviation as the task response speed feature.

7. The AGV travel path control method considering energy consumption optimization according to claim 6, characterized by, The energy consumption optimization-AGV display model is constructed according to each group of first energy consumption optimization control set and the corresponding generated optimization evaluation value, specifically: Each energy consumption optimization control set and its corresponding optimization evaluation value are one-to-one corresponding to form a mapping relationship between energy consumption optimization control set and optimization evaluation value, i.e., a point. The energy consumption optimization-AGV display model is a multi-dimensional parameter space for intuitively presenting the distribution of different control sets in energy consumption and performance. The mapping relationship between several energy consumption optimization control sets and optimization evaluation values is mapped to the energy consumption optimization-AGV display model, and the several points are fitted. The energy consumption optimization control set corresponding to the peak value is taken as the second energy consumption optimization control set and applied to the AGV path control.

8. A system using the AGV travel path control method with consideration of energy consumption optimization according to any one of claims 1 to 7, characterized by It includes an acceleration limit initialization module, a maximum load configuration initialization module, an energy consumption optimization control set collection module, a feature extraction module, and a screening control module. The acceleration limit initialization module is used to obtain a first path, initialize the acceleration limit based on the curve characteristics of the first path, and obtain a first acceleration limit. The maximum load configuration initialization module is used to estimate the maximum load according to the terrain slope and key node load demand of the first path to obtain a first maximum load configuration. The energy consumption optimization control set collection module is used to define the energy consumption optimization control set, which includes a maximum load configuration and an acceleration limit. The first maximum load configuration and the first acceleration limit are segmented and scaled, and a plurality of first energy consumption optimization control sets are generated by random combination. The feature extraction module is configured to control the AGV to run according to the first energy consumption optimization control set, extract features of each group of AGV running processes, and evaluate to generate an optimization evaluation value; The screening control module is configured to construct an energy consumption optimization-AGV display model according to each group of the first energy consumption optimization control set and the corresponding generated optimization evaluation value, perform data analysis, and screen out a second energy consumption optimization control set for AGV running path control.

9. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the AGV running path control method considering energy consumption optimization according to any one of claims 1 to 7.

10. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the AGV running path control method considering energy consumption optimization according to any one of claims 1 to 7.

Images