Single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration

Abstract

The invention discloses a single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration. The single-steering-wheel AGV path tracking method comprises the following steps of 1, dynamically adjusting the angle of an AGV steering wheel based on a preview distance; and 2, controlling the AGV to smoothly accelerate and decelerate based on a flexible acceleration and deceleration control method. The flexible acceleration and deceleration control strategy is introduced into path tracking, the acceleration and deceleration can be dynamically adjusted according to the path characteristics, the obstacle information and the current state of the AGV, and the AGV can run more stably.

Description

Technical Field

[0001] The present invention relates to the technical field of AGV control, and in particular to a single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration. Background Art

[0002] Automated Guided Vehicle (AGV) is an automated transportation system widely used in manufacturing and logistics industries for the automatic handling of materials and goods. AGV can move automatically along a predetermined path without manual operation, greatly improving logistics efficiency and safety.

[0003] Path tracking is one of the core technologies in AGV control technology, which refers to the ability of AGV to accurately and reliably travel along a predetermined path. This not only involves the positioning and navigation technology of AGV, but also includes complex control algorithms to cope with challenges such as path changes, obstacle avoidance and dynamic environment adaptation. Single-steering wheel AGV refers to AGV that uses a single steering wheel for guidance, which is widely used in automated logistics and manufacturing systems. The entire mechanical structure of the single-steering wheel AGV is simpler and more compact than the multi-steering wheel design and has extremely high flexibility. This design simplifies the mechanical structure, but puts higher requirements on the path tracking algorithm and control strategy, especially the steering and speed adjustment of the vehicle body.

[0004] The pure tracking algorithm is a path tracking control strategy commonly used in automated vehicle and robot navigation. By tracking a target point on the path, the deviation between the AGV and the predetermined path is calculated, and the driving direction of the AGV is dynamically adjusted to reduce the deviation. There are currently many technologies that combine the pure tracking algorithm.

[0005] The basic pure tracking algorithm calculates the angular difference between the vehicle's current position and the look-ahead point on the path, and adjusts the steering angle to move the vehicle toward the look-ahead point. This algorithm may not perform well when handling high speeds or sharp turns because it does not consider the impact of vehicle dynamics, such as speed and acceleration, which may cause instability or deviation in path tracking.

[0006] The pure tracking algorithm combined with fuzzy logic adjusts the steering control rules by introducing a fuzzy logic controller, which enables the algorithm to better handle uncertainty and ambiguity, such as navigation in complex environments. The design and debugging of fuzzy logic controllers are relatively complex, and the need to optimize rules and membership functions increases the difficulty of implementation.

[0007] By combining the model predictive control strategy with the pure tracking algorithm, the dynamic model of the vehicle and the path information in the future are considered to perform optimal control. Although this method can provide good path tracking performance, it has a high computational load and requires high real-time performance and computing resources, and is not suitable for single-steering wheel AGVs with limited computing power.

[0008] Using machine learning methods to learn path tracking strategies directly from data or to optimize the parameters of traditional pure tracking algorithms requires a large amount of labeled data for training, and the model has poor interpretability. When encountering new scenarios that are not covered in the training data, the performance will drop significantly.

[0009] Therefore, it is necessary to provide an improved single-steering wheel AGV path tracking method to solve the above problems. Summary of the invention

[0010] The purpose of the present invention is to provide a single-steering wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration. By introducing a flexible acceleration and deceleration control strategy in path tracking, the acceleration and deceleration can be dynamically adjusted according to the path characteristics, obstacle information and the current state of the AGV, so that the AGV can travel more smoothly.

[0011] The technical solution of the present invention is: a single-steering wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration, comprising the following steps:

[0012] Step 1: Dynamically adjust the angle of the AGV steering wheel based on the preview distance;

[0013] Step 2: Control the AGV to smoothly accelerate and decelerate based on the flexible acceleration and deceleration control method.

[0014] In the above technical solution, in step 1, the expected steering angle of the single steering wheel is calculated according to the preview distance, the expected path and the AGV state.

[0015] In the above technical solution, according to the Ackerman steering principle, the desired steering angle is:

[0016]

[0017] Among them, δ represents the expected steering angle, l represents the AGV wheelbase, L represents the preview distance, and α represents the preview angle between the current direction of the AGV and the target point.

[0018] In the above technical solution, the preview distance L satisfies the following formula:

[0019]

[0020] Where R represents the turning radius of the AGV.

[0021] In the above technical solution, in step 2, the acceleration and deceleration of the AGV is divided into 7 time periods.

[0022] In the above technical solution, in step 2, by determining the maximum speed v of the system max , maximum acceleration a max And the acceleration J determine the entire flexible acceleration and deceleration motion process.

[0023] In the above technical solution, the maximum speed v max , maximum acceleration a max And the acceleration J satisfies:

[0024]

[0025] Among them, T f represents the total running time, S o Represents the total distance covered under the overall running time.

[0026] The advantages of the present invention are:

[0027] The present invention introduces a flexible acceleration and deceleration control strategy in path tracking, which can dynamically adjust the acceleration and deceleration according to the path characteristics, obstacle information and the current state of the AGV, so as to make the AGV travel more smoothly. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0029] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention.

[0030] Figure 2 Schematic diagram of a pure tracking algorithm according to an embodiment of the present invention.

[0031] Figure 3 It is a schematic diagram comparing flexible acceleration and deceleration and T-shaped acceleration and deceleration in an embodiment of the present invention.

[0032] Figure 4 The figure is a schematic diagram of the trajectory of the AGV in an actual operation task according to an embodiment of the present invention.

[0033] Figure 5 Schematic diagram of laser AGV cross-laser sight positioning according to an embodiment of the present invention.

[0034] Figure 6 Schematic diagram of the repeatability of the positioning accuracy of the AGV working point according to an embodiment of the present invention.

[0035] Among them: 1. AGV; 2. Fork teeth; 3. Positioning laser board; 4. Cross laser line; 5. Cross laser pen. DETAILED DESCRIPTION

[0036] Example:

[0037] See also Figure 1 As shown, the present invention provides a single-steering wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration, comprising the following steps:

[0038] Step 1: Dynamically adjust the angle of the AGV steering wheel based on the preview distance;

[0039] Step 2: Control the AGV to smoothly accelerate and decelerate based on the flexible acceleration and deceleration control method.

[0040] In this embodiment, before step one, it also includes planning the task path, and after step two, it also includes repeated positioning of the AGV working point.

[0041] In this embodiment, the pure tracking algorithm focuses on the robustness and accuracy of the point-to-point working point, and adjusts the steering angle to track the predetermined path by calculating the geometric relationship between the current AGV position and the target point on the path, see Figure 2 As shown, in this embodiment, the AGV is simplified into a single vehicle model A according to the Ackerman steering geometry model.

[0042] Take the AGV control reference point P o As the current vehicle position, the preview distance between the current position and the target point G is L, the preview angle between the current direction and the target point is α, and the turning radius of the vehicle is R. According to the geometric relationship, we can get:

[0043]

[0044] When the target point is on the left side of the AGV, α>0, otherwise α<0. The bicycle model moves at an appropriate steering angle δ so that the rear axle center P of the AGV o Pass through the target point G. According to the Ackerman steering principle, the expected steering angle δ is:

[0045]

[0046] Where l represents the wheelbase of the AGV.

[0047] Define that the current AGV has a lateral deviation e relative to the target point G y , combined with the preview distance L, we can get:

[0048]

[0049] It is usually expected that the steering angle is a small angle, so the lateral deviation e at time t is y It can be approximated by the following formula:

[0050]

[0051] Where δ(t) is the desired steering angle at time t. The lateral error e y It is positively correlated with the expected steering angle. The actual AGV tracking effect is greatly affected by the preview distance L. The preview distance L represents the longitudinal linear speed of the AGV, that is:

[0052] L=λv x +c

[0053] Where λ is the proportionality coefficient, and c is the basic preview distance which is a constant. In general, the preview distance L will increase with the increase of vehicle speed. This is because the AGV needs a longer reaction distance to adapt to the change of the path when driving at high speed. The present invention takes into account the safety of the interactive environment and other factors, and sets the maximum speed v of the AGV to max The speed is limited to 1.2 m / s, so λ is set to 0.87 and c is set to 0.3. Different adjustment parameters need to be set in different application scenarios.

[0054] The traditional linear acceleration and deceleration method will cause great impact and vibration to the mechanical parts, which is not conducive to the long-term use of the AGV and will cause mechanical wear. The gradual acceleration and deceleration method allows the AGV to start and stop smoothly, improves the stability of the system movement, and provides better working conditions for long-term high-precision positioning and robust control. Flexible acceleration and deceleration is different from traditional T-shaped acceleration. Acceleration and deceleration are divided into 7 time periods. See Figure 3 shown.

[0055] Line I in the figure is the speed curve of flexible acceleration and deceleration, line II is the traditional T-shaped acceleration and deceleration speed curve, and line III is the relationship between the acceleration and deceleration of the flexible curve and time. k =t k -t k-1 (k=1,2,...,7). The acceleration a during flexible acceleration and deceleration will change with the progress. The given variable acceleration J is:

[0056]

[0057] According to the 7-stage acceleration and deceleration, the relationship between acceleration and time can be obtained as follows:

[0058]

[0059] The velocity and acceleration satisfy the relationship v = at, and the velocity and jerk satisfy the relationship v = 1 / 2 (Jt 2 ), the relationship between displacement S and time can be deduced as:

[0060]

[0061] Thus, the maximum speed v of the system is determinedmax , maximum acceleration a max The entire flexible acceleration and deceleration process can be determined by the jerk J. The jerk J reflects the overall flexibility of the system. The greater the flexibility, the greater the overshoot and the shorter the running time.

[0062] In a complete flexible acceleration and deceleration process T f The maximum speed v needs to be reached during operation max , and in an ideal state, T1=T3=T5=T7. According to the relationship between time process, speed and acceleration, if T2 exists, then according to the relationship between speed and time in the above formula, the maximum speed v max for:

[0063]

[0064] A max =JT1, the expression of T2 is:

[0065]

[0066] Where v0 is the initial speed of the AGV. Given the total running time T f In the case of , T4 can be derived as:

[0067]

[0068] The jerk J can be calculated from the total running time T f The total distance S o get:

[0069]

[0070] According to the process of controlling the jerk in the acceleration and deceleration stages by the flexible acceleration and deceleration, the influence of the speed mutation can be reduced to achieve a smooth vehicle speed control effect.

[0071] Based on the above technical solution, the present invention has the following beneficial effects:

[0072] 1. Improve the path tracking accuracy of single-steering wheel AGV

[0073] In order to meet the high requirements for the accuracy of AGV's pick-up and delivery points in work tasks, the present invention adopts a pure tracking algorithm based on the preview distance, which can ensure that the single-steering wheel AGV closely follows the predetermined path and reduce the path deviation by dynamically adjusting the angle of the steering wheel. This method is particularly suitable for situations where the path contains many bends, which can significantly reduce the path tracking error and improve the navigation accuracy.

[0074] 2. Improve the speed smoothness of vehicle acceleration and deceleration

[0075] In order to solve the problem of sudden speed changes in the traditional T-shaped acceleration and deceleration method, the present invention uses an S-shaped flexible adjustment of the 7-stage acceleration process to ensure that the AGV can operate efficiently and safely in different scenarios. By adopting a gradual acceleration and deceleration strategy, the AGV can achieve a smoother start and stop process. The present invention not only enhances the overall motion stability of the system, but also creates more favorable conditions for continuous high-precision positioning and powerful robustness control. Without sacrificing the response speed, it can effectively reduce the vibration and impact that may occur during the start and stop process, thereby extending the service life of the AGV and its carrying system.

[0076] 3. Improve the repeatability of the single-steering-wheel AGV working point

[0077] The present invention pays more attention to the point-to-point accuracy in the work task, especially the repeated positioning accuracy of the work point. The experiment verifies that the present invention has high accuracy and robustness in the indoor structured scene of the logistics industry. By making full use of these advantages, a more flexible, reliable and efficient AGV system can be realized to meet the high standard requirements of modern industry for automated transportation and material handling.

[0078] The solution of the present invention is verified by experiments below.

[0079] The experimental test was conducted on a modified forklift, with the maximum vehicle speed limited to 1.2m / s, the maximum preview distance 3m, and the minimum preview distance 0.3m. This invention focuses on the long-term AGV delivery task based on the actual application needs of the structured scene in the smart logistics room. P is the set work station. See Figure 4 As shown in the figure, there are three motion trajectories of the AGV work task.

[0080] Depend on Figure 4 It can be seen that the AGV of the present invention can obtain accurate positioning in real time, and can obtain smooth motion trajectories by repeating multiple times in the pure tracking algorithm and the flexible acceleration and deceleration algorithm. According to the partial enlarged diagram, it can be seen that the present invention obtains robust results in straight and curved motion. Figure 4 (d) is the starting site. When the initial position is uncertain, the AGV will quickly fit the route, and the results of multiple movements in subsequent curves and straight lines tend to be stable and have high robustness.

[0081] According to the annotation information of the site, the motion trajectory is divided into linear motion and curved motion, and the repeated positioning accuracy of the working point is verified in the next section. In order to better verify the robustness of trajectory tracking, Table 1 selects 3 trajectory measurements and intercepts the positioning accuracy analysis of the linear motion trajectory. The unit is meter (m), and the RMSE in the table is the root mean square error.

[0082] Table 1 Analysis of planned and actual paths of linear motion

[0083]

[0084] According to the data in Table 1, the root mean square error of the path tracking algorithm of the single steering wheel AGV of the present invention is within 25 mm, indicating that the straight motion path tracking effect of the forklift of the present invention is good. Table 2 shows the curve motion path tracking error analysis from P2 to P3.

[0085] Table 2 Analysis of planned path and actual path of curvilinear motion

[0086]

[0087] According to the data in Table 2, the curve motion trajectory error under three trajectory tracking is obtained, and the root mean square error of path tracking is within 35 mm, which verifies that the algorithm of the present invention can efficiently and safely perform path tracking in a structured factory area.

[0088] The two-dimensional laser single-steering wheel AGV method proposed in the present invention pays more attention to the task of accurately arriving at the delivery point, and the repeatability of the work station can directly reflect whether the parking position of the AGV meets expectations.

[0089] See also Figure 5 As shown, in order to accurately measure the repeat site positioning accuracy of AGV 1 in actual work tasks, the present invention fixes a cross laser pen 5 on one side of the fork 2 through a mechanical firmware, and projects a cross laser line 4 onto a positioning laser plate 3 on the ground to determine the position of AGV 1.

[0090] The repeatability of the workstation can be determined by the position of the laser crosshair on the cross plate. The present invention records the results of 120 workstation positioning. Figure 6 The precision unit of the dots shown is millimeters (mm).

[0091] Continue to see Figure 6 As shown in the figure, the working point repositioning of the AGV of the present invention can be completely wrapped by a circle with a radius of 6mm, which shows that the working point repositioning accuracy of the algorithm of the present invention can reach ±6mm. After multiple actual verifications, it is proved that the single-steering wheel AGV forklift algorithm based on two-dimensional laser proposed in the present invention has high accuracy and robustness in the indoor structured scenes of the logistics industry.

[0092] Of course, the above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any modifications made according to the spirit of the main technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration, characterized in that: The following steps are involved: Step 1: Dynamically adjust the angle of the AGV steering wheel based on the preview distance; Step 2: Control the AGV to smoothly accelerate and decelerate based on the flexible acceleration and deceleration control method.

2. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 1 is characterized in that: In the step 1, the expected steering angle of the single steering wheel is calculated according to the preview distance, the expected path and the AGV state.

3. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 2 is characterized in that: According to the Ackerman steering principle, the desired steering angle is: Among them, δ represents the expected steering angle, l represents the AGV wheelbase, L represents the preview distance, and α represents the preview angle between the current direction of the AGV and the target point.

4. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 3 is characterized in that: The preview distance L satisfies the following formula: Where R represents the turning radius of the AGV.

5. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 1 is characterized in that: In step 2, the acceleration and deceleration of the AGV is divided into 7 time periods.

6. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 1 is characterized in that: In step 2, by determining the maximum speed v of the system max , maximum acceleration a max And the acceleration J determine the entire flexible acceleration and deceleration motion process.

7. The single-steering-wheel AGV path tracking method based on pure tracking and flexible acceleration and deceleration according to claim 6 is characterized in that: The maximum speed v max , maximum acceleration a max And the acceleration J satisfies: Among them, T f represents the total running time, S o Represents the total distance covered under the overall running time.

Images