AGV intelligent control system oriented to complex dynamic environment

Abstract

The invention provides an AGV intelligent control system for a complex dynamic environment, and the system comprises a dynamic map module which is used for obtaining the dynamic level of each perception factor through processing according to the real-time heterogeneous information collected by a multi-fusion sensor, thereby constructing a hierarchical dynamic map, and each perception factor in the map has a dynamic attribute; the dynamic attribute comprises a dynamic level; stabilizing the confidence coefficient; triggering a priority mark; the path planning and reconstruction module is used for judging whether a current main path needs to be reconstructed according to information provided by the hierarchical dynamic map, calling a TEB algorithm when the current main path needs to be reconstructed, increasing virtual cost in a high dynamic weight region in a path cost map, and adding a path constraint term in a TEB optimization objective function, so that the path planning and reconstruction of the current main path is realized. The path constraint term is related to the dynamic attribute; after the path is generated, stability verification is carried out, if verification is not passed, the result is fed back to the dynamic map module, a map area is requested to be expanded, and retry is carried out.

Description

Technical Field

[0001] The present invention relates to the field of AGV technology, and in particular to an AGV intelligent control system for complex dynamic environments. Background Art

[0002] Existing AGVs (Automated Guided Vehicles) are widely used in warehousing, logistics, manufacturing, and medical transportation, primarily relying on fixed routes or static maps for navigation. However, in real-world scenarios, AGVs often face dynamic environmental disturbances, such as new obstacles (such as pedestrians and temporarily stored items), changes in surface conditions (slippery, vibrating, and pothole-prone), and adjustments to task scheduling. Traditional controllers lack sufficient flexibility to handle these changes, which can easily lead to path planning failures, task interruptions, and even collisions.

[0003] Therefore, how to enable AGV to operate stably in complex and real-time changing scenarios is a problem that needs to be solved at present. Summary of the Invention

[0004] The purpose of this invention is to propose an AGV intelligent control system for complex dynamic environments, which can enable AGV to operate stably in complex and real-time changing scenarios.

[0005] To achieve the above objectives, the present invention provides an AGV intelligent control system for complex dynamic environments, comprising: The dynamic map module is used to process the real-time heterogeneous information collected by multiple fusion sensors to obtain the dynamic level of each perception factor, thereby constructing a hierarchical dynamic map, and each perception factor in the map has dynamic attributes; The dynamic attributes include: Dynamic level, used to characterize the probability of change of perception factors; Stable confidence, used to characterize the continuity of perception factors within the time window; Trigger priority flag, used to indicate whether to trigger path reconstruction or speed adjustment; The path planning and reconstruction module determines whether the current main path needs to be reconstructed based on the information provided by the hierarchical dynamic map. When the path needs to be reconstructed, the TEB algorithm is called to increase the virtual cost of the area with high dynamic weight in the path cost map, and add a path constraint item related to the dynamic attribute to the TEB optimization objective function. After the path is generated, a stability check is performed. If the check fails, feedback is given to the dynamic map module to request the expansion of the map area and retry.

[0006] In the optional solution, the dynamic levels are divided into: stable factor, semi-dynamic factor, and high dynamic factor.

[0007] In an optional solution, path reconstruction is required when the information provided by the hierarchical dynamic map meets any of the following conditions: The obstacle high variation index path point coverage rate is greater than the first threshold; The perception factor with high trigger priority is at the front of the path Inside; Stable confidence Frequent fluctuations.

[0008] In the optional solution, the virtual cost is: in: It is the base cost value in the regular cost map; is the dynamic weight amplification coefficient; is the dynamic weight coefficient, which indicates the probability that the position (x, y) is disturbed by an obstacle.

[0009] In the optional scheme, the optimized objective function is: in, is the traditional TEB cost; is the weight coefficient of the VPI cost item; in, is the i-th candidate path point on the path; Dj(Xi) represents the distance between the i-th path point on the path and the nearest obstacle; To stabilize confidence; To trigger the priority; is the dynamic weight index; is the time since the last appearance; is the time decay coefficient.

[0010] In the optional solution, the stability check items include at least one of the following three: Whether the proportion of high dynamic areas in the path segment is less than the safety threshold; Whether the path maintains a safe margin between the known high dynamic factors in the map; Whether the start and end points are aligned with the current pose / target point.

[0011] In an optional solution, the system further includes a DOB module, a self-tuning module, a PID controller and a control actuator; The DOB module senses external disturbances in real time, calculates the error between the actual pose and the target trajectory, and estimates the total disturbance in real time; The DOB module outputs the total disturbance change rate, error and error change rate to the self-tuning module; The self-tuning module adjusts the PID parameters in real time according to the output of the DOB module and the load status; The PID controller acts on the control actuator according to the PID parameters adjusted in real time. The control actuator forms a final input according to the total disturbance output by the DOB module and the output of the PID controller, and feeds it back to the DOB module, so that the DOB module compensates for the external disturbance.

[0012] In the optional solution, the total disturbance calculation formula is: is a low-pass filter; is the system nominal transfer function; is the actual posture; is the final input.

[0013] In an optional solution, adjusting the PID parameters includes: Small error but oscillation: Increase; There is a steady-state error: Increase; Total disturbance changes rapidly: temporary Increase; Load increase: , Increase.

[0014] The beneficial effects of the present invention are: The control system of the present invention has the ability to perceive dynamic environments and reconstruct paths, and can achieve stable operation in complex and real-time changing scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings, in which like reference numerals generally represent like components.

[0016] Figure 1 This is an architectural diagram of an AGV intelligent control system for complex dynamic environments in one embodiment of the present invention.

[0017] Figure 2 This is a disturbance closed-loop adaptive control mechanism in one embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and drawings. However, it should be noted that the technical solutions of the present invention can be implemented in a variety of different forms and are not limited to the specific embodiments described herein. The drawings are highly simplified and not to exact scale, and are intended solely to facilitate and clearly illustrate the embodiments of the present invention.

[0019] It should be understood that when an element or layer is referred to as being "on," "adjacent to," "connected to," or "coupled to" another element or layer, it can be directly on, adjacent to, connected to, or coupled to the other element or layer, or there can be intervening elements or layers. Conversely, when an element is referred to as being "directly on," "directly adjacent to," "directly connected to," or "directly coupled to" another element or layer, there are no intervening elements or layers. It should be understood that although the terms first, second, third, etc. may be used to describe various elements, components, regions, layers, and / or parts, these elements, components, regions, layers, and / or parts should not be limited by these terms. These terms are merely used to distinguish one element, component, region, layer, or part from another element, component, region, layer, or part. Thus, a first element, component, region, layer, or part discussed below may be represented as a second element, component, region, layer, or part without departing from the teachings of the present invention.

[0020] Spatially relative terms such as "under," "beneath," "below," "under," "above," "above," etc., may be used herein for convenience of description to describe the relationship of one element or feature shown in the figures to other elements or features. It should be understood that the spatially relative terms are intended to include different orientations of the device in use and operation in addition to the orientations shown in the figures. For example, if the device in the drawings is flipped, then the elements or features described as "under" or "beneath" or "beneath" the other elements will be oriented as "over" the other elements or features. Thus, the exemplary terms "under" and "under" may include both the upper and lower orientations. The device may be oriented otherwise (rotated 90 degrees or in other orientations) and the spatial descriptors used herein are interpreted accordingly.

[0021] The terminology used herein is intended only to describe specific embodiments and is not intended to limit the present invention. When used herein, the singular forms "a," "an," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "comprising" and / or "including," when used in this specification, identify the presence of the recited features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0022] Example 1 Reference Figure 1 and Figure 2 This embodiment provides an AGV intelligent control system for complex dynamic environments, including: The dynamic map module is used to process the real-time heterogeneous information collected by multiple fusion sensors to obtain the dynamic level of each perception factor, thereby constructing a hierarchical dynamic map, and each perception factor in the map has dynamic attributes; The dynamic attributes include: Dynamic Level , used to characterize the probability of change of perception factors; Stable confidence , used to characterize the continuity of the perception factor within the time window; Trigger priority tag , used to characterize whether path reconstruction or speed adjustment is induced; The path planning and reconstruction module determines whether the current main path needs to be reconstructed based on the information provided by the hierarchical dynamic map. When the path needs to be reconstructed, the TEB algorithm is called to increase the virtual cost of the area with high dynamic weight in the path cost map, and add a path constraint item related to the dynamic attribute to the TEB optimization objective function. After the path is generated, a stability check is performed. If the check fails, feedback is given to the dynamic map module to request the expansion of the map area and retry.

[0023] Specifically, the traditional AGV perception layer focuses only on obstacle detection and basic map construction. This embodiment innovatively introduces a "hierarchical dynamic map" based on this. By integrating heterogeneous information collected by the LiDAR + Visual SLAM + Infrared Array module, each factor category is dynamically updated during map construction, and a hierarchical dynamic map is constructed. Each factor in the map not only records the spatial location, but also has the following dynamic attributes: (1) Dynamic level (Reflecting its probability of change); (2) Stable confidence (Reflects the continuity of the target within the time window); (3) Trigger priority mark (whether to trigger path reconstruction or speed adjustment).

[0024] Referring to Table 1, perception factors are divided into three categories according to their dynamic characteristics: Table 1 The hierarchical dynamic map provides the following three trigger signals to determine whether the current main path needs to be reconstructed: If the path point coverage rate of the high obstacle variation index is ≥ the first threshold, it means that the path passes through an area that is likely to be disturbed. The perception factor with high trigger priority is at the front of the path (unit: meter); Stable confidence Frequent fluctuations: if there are two consecutive drops, the area is considered no longer accessible; If any of the conditions are met, the path reconstruction phase begins.

[0025] When reconstructing the path, the Timed Elastic Band (TEB) algorithm is called and the following perceptual graph guidance information is added: Costmap update: In the path cost map, the area with high dynamic weight is increased by virtual cost Cost(x,y): in: It is the base cost value in the regular cost map; Dynamic weight amplification factor; is the dynamic weight coefficient, which indicates the probability that the position (x, y) is disturbed by an obstacle.

[0026] Add path constraint terms to the TEB optimization objective function: Add the VPI term (path constraint term) to the TEB optimization objective function to make the path actively avoid the "hotspot dynamic area". in: is the traditional TEB cost; is the weight coefficient of the VPI cost term, which is used to adjust the importance of predicting obstacle avoidance; in, is the i-th candidate path point on the path; Dj(Xi) represents the distance between the i-th path point on the path and the nearest obstacle; To stabilize confidence; To trigger the priority; is the dynamic weight index; is the time since the last appearance; is the time decay coefficient. VPI is not only based on the obstacle movement history, but also considers the obstacle type and stability confidence.

[0027] After the path is generated, the following checks are performed: Is the proportion of high-dynamic areas within the path segment less than a safety threshold? Is there a safety margin (e.g., ≥ 0.5 meters) between the path and known high-frequency disturbance factors in the map? Is the start and end points accurately aligned with the current pose / target point? If these conditions are not met, feedback is sent to the perception layer to request expansion of the map area and retry.

[0028] Traditional AGV controllers (such as fixed-parameter PID / MPC) suffer from parameter rigidity and lack of active anti-disturbance capabilities when faced with dynamic disturbances (such as slope, load fluctuations, and changes in ground friction). This leads to reduced trajectory tracking accuracy and unstable operation. Therefore, this embodiment proposes a disturbance-tolerant closed-loop adaptive control mechanism.

[0029] The system also includes a DOB (disturbance observer) module, a self-tuning module, a PID controller and a control actuator; the DOB module senses external disturbances in real time, calculates the error between the actual posture and the target trajectory, and estimates the total disturbance in real time; the DOB module outputs the total disturbance change rate, the error and the error change rate to the self-tuning module; the self-tuning module adjusts the PID parameters in real time according to the output of the DOB module and the load status; the PID controller acts on the control actuator according to the PID parameters adjusted in real time, and the control actuator forms a final input based on the total disturbance output by the DOB module and the output of the PID controller, and feeds it back to the DOB module, so that the DOB module compensates for the external disturbance.

[0030] The DOB module senses and compensates for external disturbances (such as slope, load changes, and minor collisions) in real time, integrates IMU and encoder data, and calculates the error between the actual pose and the target trajectory.

[0031] The total disturbance calculation formula is: .

[0032] It is a low-pass filter that suppresses noise and ensures stability; is the system nominal transfer function; is the actual pose, obtained by each sensor; For the final input, . Actively eliminate the effects of disturbances.

[0033] The self-tuning module introduces self-adjustment rules, combining the control error trend, disturbance change rate, and load estimation value to adjust the PID parameters in real time. The following rules are used: Large error and rate of change: significant undershoot / overshoot, Increase, enhance the corresponding Increase to suppress overshoot.

[0034] Small error but oscillation: insufficient damping, Increase.

[0035] There is a steady-state error: insufficient integration, Increase.

[0036] Total disturbance (k) Rapid changes: severe disturbances, temporary Increase, enhance anti-interference.

[0037] Load increases: inertia increases, , Increase, and vice versa.

[0038] The self-tuned PID parameters act on the system, and the new control effect (error, error change rate) generated is fed back to the DOB module and the self-tuning module for disturbance estimation and parameter optimization in the next cycle.

[0039] This embodiment introduces multi-level perception fusion, dynamic map reconstruction, adaptive control, and interference feedback mechanisms to form a closed-loop system of "active perception-prediction reconstruction-control coordination", which can significantly improve the autonomous operation capability, path planning flexibility, and control robustness of AGVs in unstructured environments. The beneficial effects are as follows: 1. Improved adaptability to environmental dynamics, accurate perception of highly dynamic targets, significantly reduced probability of path planning failure, proactive avoidance of high-risk areas, and improved path safety and collision avoidance capabilities.

[0040] 2. Enhanced control robustness effectively suppresses disturbances such as slope changes and load fluctuations, improves trajectory tracking accuracy, and adaptively adjusts parameters to enhance operational stability under conditions such as high-speed emergency stops and sudden load changes.

[0041] 3. Universal expansion of complex scenarios, reducing reliance on manual parameter adjustment, increasing system response speed, supporting unstructured environments (such as temporary warehouses), adapting to dynamic scenarios in multiple industries, and reducing deployment and adaptation costs and time.

[0042] The above description is only a description of the preferred embodiments of the present invention and does not limit the scope of the present invention. Any changes and modifications made by ordinary technicians in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. An AGV intelligent control system for complex dynamic environments, characterized by: include: The dynamic map module is used to process the real-time heterogeneous information collected by multiple fusion sensors to obtain the dynamic level of each perception factor, thereby constructing a hierarchical dynamic map, and each perception factor in the map has dynamic attributes; The dynamic attributes include: Dynamic level, used to characterize the probability of change of perception factors; Stable confidence, used to characterize the continuity of perception factors within the time window; Trigger priority flag, used to indicate whether to trigger path reconstruction or speed adjustment; The path planning and reconstruction module determines whether the current main path needs to be reconstructed based on the information provided by the hierarchical dynamic map. When the path needs to be reconstructed, the TEB algorithm is called to increase the virtual cost of the area with high dynamic weight in the path cost map, and add a path constraint item related to the dynamic attribute to the TEB optimization objective function. After the path is generated, a stability check is performed. If the check fails, feedback is given to the dynamic map module to request the expansion of the map area and retry.

2. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: The dynamic levels are divided into: stable factor, semi-dynamic factor, and high dynamic factor.

3. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: When the information provided by the hierarchical dynamic map meets any of the following conditions, path reconstruction is required: The obstacle high variation index path point coverage rate is greater than the first threshold; The perception factor with high trigger priority is at the front of the path Inside; Stable confidence Frequent fluctuations.

4. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: The virtual cost is: in: It is the base cost value in the regular cost map; is the dynamic weight amplification coefficient; is the dynamic weight coefficient, indicating the position ( ) is interfered by obstacles.

5. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: The optimized objective function is: in, is the traditional TEB cost; is the weight coefficient of the VPI cost item; in, is the i-th candidate path point on the path; Dj(Xi) represents the distance between the i-th path point on the path and the nearest obstacle; To stabilize confidence; To trigger the priority; is the dynamic weight index; is the time since the last appearance; is the time decay coefficient.

6. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: The stability check items include at least one of the following three: Whether the proportion of high dynamic areas in the path segment is less than the safety threshold; Whether the path maintains a safe margin between the known high dynamic factors in the map; Whether the start and end points are aligned with the current pose / target point.

7. The AGV intelligent control system for complex dynamic environments according to claim 1, characterized in that: The system also includes a DOB module, a self-tuning module, a PID controller and a control actuator; The DOB module senses external disturbances in real time, calculates the error between the actual pose and the target trajectory, and estimates the total disturbance in real time; The DOB module outputs the total disturbance change rate, error and error change rate to the self-tuning module; The self-tuning module adjusts the PID parameters in real time according to the output of the DOB module and the load status; The PID controller acts on the control actuator according to the PID parameters adjusted in real time. The control actuator forms a final input according to the total disturbance output by the DOB module and the output of the PID controller, and feeds it back to the DOB module, so that the DOB module compensates for the external disturbance.

8. The AGV intelligent control system for complex dynamic environments according to claim 7, characterized in that: The total disturbance calculation formula is: is a low-pass filter; is the system nominal transfer function; is the actual posture; is the final input.

9. The AGV intelligent control system for complex dynamic environments according to claim 7, characterized in that: The adjusting of PID parameters comprises: Small error but oscillation: Increase; There is a steady-state error: Increase; Total disturbance changes rapidly: temporary Increase; Load increase: , Increase.

Images