Unmanned agricultural machine dynamic obstacle avoidance method, system and device based on guide vector field

Abstract

The invention discloses an unmanned agricultural machine dynamic obstacle avoidance method, system and device based on a guide vector field, and the method comprises the steps: building an unmanned agricultural machine kinematic model containing external disturbance according to the motion characteristics of an unmanned agricultural machine; combining the attraction vector field and the rejection vector field to form a guide vector field; based on the kinematic model and the guide vector field, designing a disturbance observer, and estimating external disturbance in real time; based on external disturbance, a non-smooth obstacle avoidance controller is designed and used for adjusting the front wheel rotation angle of the unmanned agricultural machine, tracking the expected course angle of the guide vector field and achieving dynamic obstacle avoidance; when the course error of the unmanned agricultural machine is large, the unmanned agricultural machine actively reduces the speed, so that the agricultural machine drive-by-wire chassis has enough time to execute steering operation. The problem of control deviation caused by external disturbance is effectively solved, and quick response and high-precision control performance of the system are guaranteed; in addition, the obstacle avoidance capability of the agricultural machine in a dynamic environment is remarkably improved, and then the operation quality is optimized.

Description

Technical Field

[0001] The present invention belongs to the technical field of agricultural machinery control, and particularly relates to a dynamic obstacle avoidance method and system for unmanned agricultural machinery based on a guidance vector field. Background Art

[0002] Under the dual pressures of the continuous growth of the global population and the increasing shortage of agricultural labor, the agricultural field has an increasingly urgent need for efficient and precise unmanned agricultural machinery technology. The operation of traditional agricultural machinery in complex dynamic environments not only relies on a large amount of manpower but also has low efficiency, making it difficult to meet the modern requirements of precision agriculture. In recent years, in response to the challenges of food security and labor shortage, many countries around the world are actively promoting the research and development and application of unmanned agricultural machinery. These machines have significantly reduced the manpower requirements in agricultural production by integrating advanced autonomous navigation and control technologies and have become the core support for realizing precision agriculture. However, although significant progress has been made in the autonomous navigation of unmanned agricultural machinery technology, its dynamic obstacle avoidance ability still faces severe challenges. Especially when dealing with dynamic obstacles and external disturbances in complex environments, the accuracy and stability of the system need to be improved urgently. Summary of the Invention

[0003] Aiming at the deficiencies in the prior art, the present invention provides a dynamic obstacle avoidance control method and system for unmanned agricultural machinery based on a guidance vector field, which is applicable to a variety of agricultural equipment. The system ensures that unmanned agricultural machinery can stably avoid obstacles and accurately track preset targets in complex environments by constructing a dynamic model containing unknown disturbances, designing a guidance vector field, developing a non-smooth obstacle avoidance controller, and real-time estimating and compensating external disturbances. The present invention significantly improves the obstacle avoidance performance and operation quality of agricultural machinery and provides strong support for smart agriculture.

[0004] The present invention achieves the above technical objectives through the following technical means.

[0005] A dynamic obstacle avoidance method for unmanned agricultural machinery based on a guidance vector field:

[0006] According to the motion characteristics of unmanned agricultural machinery, establish a kinematic model of unmanned agricultural machinery containing external disturbances;

[0007] Combine the attraction vector field and the repulsion vector field to form a guidance vector field;

[0008] Based on the kinematic model of unmanned agricultural machinery and the guidance vector field, design a disturbance observer to monitor and accurately estimate external disturbances in real time;

[0009] Based on the disturbance information provided by the disturbance observer, design a non-smooth obstacle avoidance controller;

[0010] The non-smooth obstacle avoidance controller adjusts the front wheel angle of the unmanned agricultural machinery to track the desired heading angle of the guidance vector field, achieving the purpose of dynamic obstacle avoidance;

[0011] The non-smooth obstacle avoidance controller is as follows:

[0012]

[0013] where δ is the front wheel steering angle of the unmanned agricultural machine, L is the wheelbase between the front and rear axles of the unmanned agricultural machine, v is the speed of the unmanned agricultural machine, s is the sliding mode variable, and s = e ψ , the heading error e ψ = ψ - ψ r , ψ is the heading angle of the unmanned agricultural machine, ψ r is the expected heading angle of the guidance vector field, η 1 , η 2 , η 3 are controller parameters, α and β are positive constants, is the estimated value of the lumped disturbance, and sign(x) is the sign function.

[0014] Furthermore, when the heading error of the unmanned agricultural machine is greater than or equal to the set threshold, the unmanned agricultural machine actively reduces its speed, so that the wire-controlled chassis of the agricultural machine has enough time to perform the steering operation, avoiding potential collision risks in practical applications:

[0015]

[0016] where v 0 is the set agricultural machine speed, s th and γ are positive constants, s th is the heading error threshold determined by experiments.

[0017] Furthermore, the disturbance observer is as follows:

[0018]

[0019] where, is the estimated value of the sliding mode variable s, and are respectively 's first-order derivatives, and are observer parameters selected artificially.

[0020] Furthermore, the guidance vector field is as follows:

[0021]

[0022] where F(r) is the guidance vector field, σ i (β i ) is the mixing function, F a (r t ) is the attracting vector field, F r，i (rs，i ) is the repulsive vector field of the i-th obstacle, r t = r - r target , r s，i = r - r o，i , r is the position of the unmanned agricultural machine, r target is the position of the target point tracked by the unmanned agricultural machine, r o，i is the position of the i-th obstacle, N is the number of obstacles, intermediate quantity β i = -||r - r s，i || 2 , constant β i = -||ρ b + ρ o，i + ρ a，i || 2 , constant ρ b is the radius of the smallest circle enclosing the unmanned agricultural machine, ρ o，i is the radius of the i-th obstacle, ρ a，i and ρ s，i are user-defined parameters, ξ(β i ) is a mixing function related to β i , and

[0023] Further, the attractive vector field

[0024] Further, the repulsive vector field of the i-th obstacle where p is an adjustable unit vector.

[0025] Further, the desired heading angle ψ of the guidance vector field r = arctan(F y / F x ), F y is the component of the guidance vector field F(r) in the y direction, F x is the component of the guidance vector field F(r) in the x direction.

[0026] Further, the kinematic model of the unmanned agricultural machine including external disturbances is: where d 1 , d 2 , d 3 respectively represent external disturbances on different channels, including the position in the x direction of the geodetic coordinate system, the position in the y direction of the geodetic coordinate system, and the heading of the agricultural machine, (x, y) represents the position coordinates of the unmanned agricultural machine in the geodetic coordinate system.

[0027] A dynamic obstacle avoidance system for unmanned agricultural machines based on a guidance vector field, comprising:

[0028] Agricultural machinery kinematic model construction module: responsible for establishing the kinematic model of unmanned agricultural machinery including external disturbances;

[0029] Guiding vector field construction module: designs an attraction vector field and a repulsion vector field, and combines the two to form a guiding vector field;

[0030] Disturbance observer design module: designs a disturbance observer based on the kinematic model of unmanned agricultural machinery and the guiding vector field to monitor and accurately estimate external disturbances in real time;

[0031] Non-smooth obstacle avoidance controller design module: constructs a non-smooth obstacle avoidance controller with disturbance compensation function based on the external disturbances provided by the disturbance observer;

[0032] Obstacle avoidance speed adaptive strategy module: realizes the active speed reduction of unmanned agricultural machinery when the heading error of the unmanned agricultural machinery is greater than or equal to the set threshold.

[0033] A dynamic obstacle avoidance device for unmanned agricultural machinery based on a guiding vector field, comprising:

[0034] Host computer module, used to send the position of the target point tracked by the unmanned agricultural machinery;

[0035] Environmental perception system, used to identify the position, shape and size of obstacles;

[0036] Navigation and positioning system, used to continuously obtain the real-time position and path of the unmanned agricultural machinery;

[0037] Digital processor unit, used to construct the kinematic model of unmanned agricultural machinery, guiding vector field, design disturbance observer and non-smooth obstacle avoidance controller;

[0038] Agricultural machinery by-wire chassis system, receives the control instructions of the digital processor unit, and controls the actions of the unmanned agricultural machinery;

[0039] Vehicle communication system, responsible for data exchange, ensuring data synchronization and collaborative work among the host computer module, environmental perception system, navigation and positioning system, digital processor unit and agricultural machinery by-wire chassis system.

[0040] The beneficial effects achieved by the present invention are as follows:

[0041] (1) The guiding vector field of the present invention combines an attraction vector field and a repulsion vector field, which can ensure that the unmanned agricultural machinery effectively avoids obstacles while maintaining accurate tracking of the target path, significantly enhancing the obstacle avoidance ability of the agricultural machinery in complex environments.

[0042] (2) The disturbance observer of the present invention is used to observe external disturbances and perform feedforward compensation; this disturbance observer can estimate external disturbances in real time and feedback them into the control loop, offsetting the impact of disturbances on the system through feedforward compensation; it not only improves the tracking accuracy of unmanned agricultural machinery in complex environments, but also enhances the robustness of the system.

[0043] (3) The non-smooth obstacle avoidance controller of the present invention enhances the anti-disturbance ability of the system by introducing an integral term and uses non-smooth control technology to quickly adjust the control input, thus significantly improving the tracking accuracy of unmanned agricultural machinery in complex dynamic environments; the non-smooth obstacle avoidance controller can effectively improve the stability and reliability of unmanned agricultural machinery operations when facing external disturbances and system uncertainties.

[0044] (4) The speed adaptive strategy of the present invention actively reduces the forward speed of unmanned agricultural machinery. When facing a large heading angle error, it can provide enough time for the non-smooth obstacle avoidance controller to perform steering operations, thus significantly reducing the potential collision risk, effectively improving the obstacle avoidance ability, and enhancing the safety of unmanned agricultural machinery in complex environments.

[0045] (5) The upper computer module of the present invention provides an intuitive operation interface, enabling operators to easily input parameters, monitor the status, and make real-time adjustments, ensuring the stable operation of the entire system, and facilitating timely problem-solving, which can reduce downtime and effectively improve the stability and reliability of operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a schematic diagram of the dynamic obstacle avoidance control of the unmanned agricultural machinery of the present invention;

[0047] Figure 2 is a schematic diagram of the composition of the dynamic obstacle avoidance device of the unmanned agricultural machinery of the present invention;

[0048] Figure 3 is the tracking and obstacle avoidance curve graph of the unmanned agricultural machinery in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The present invention provides a dynamic obstacle avoidance method and system for unmanned agricultural machinery based on a guidance vector field, applicable to various agricultural equipment. The system ensures that unmanned agricultural machinery can stably avoid obstacles and accurately track a preset target in complex environments by constructing a kinematic model of unmanned agricultural machinery containing unknown disturbances, designing a guidance vector field, developing a non-smooth obstacle avoidance controller, and real-time estimating and compensating external disturbances. The schematic diagram of its overall control scheme is as Figure 1As shown below. For the convenience of understanding, to make the purpose, technical solution and effects of the present invention clearer and more definite, the present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the described embodiments are only part of the embodiments of the present invention, not all embodiments. It is only used to explain the present invention and is not used to limit the present invention.

[0050] The composition of a dynamic obstacle avoidance device for unmanned agricultural machinery based on a guiding vector field is as Figure 2 shown, and will be used to specifically illustrate the present invention. The system of the present invention mainly includes the following modules:

[0051] Host computer module: The host computer can directly issue control commands, such as sending the position of the target point tracked by the unmanned agricultural machinery to the guiding vector field module in real time; record various data generated during the experiment, such as position information, speed, steering angle, etc., and display the working status of the agricultural machinery in real time through a graphical interface, which is convenient for operators to monitor and experiment.

[0052] Environment perception system: Equipped with sensors such as lidar, it is used to perceive the surrounding environment of the unmanned agricultural machinery in real time, and can identify the position, shape and size of obstacles, providing accurate data support for the obstacle avoidance of the unmanned agricultural machinery.

[0053] Navigation and positioning system: Combining GNSS (Global Navigation Satellite System) and IMU (Inertial Measurement Unit), it is used to continuously obtain the real-time position and path of the unmanned agricultural machinery. Through high-precision positioning technology and multi-sensor fusion technology, it ensures that the position and driving path of the agricultural machinery can be obtained in real time, contributing to precision agriculture.

[0054] Vehicle communication system: Responsible for data communication between components of the unmanned agricultural machinery, including receiving and sending necessary data, ensuring data synchronization and collaborative work between the host computer module, environment perception system, navigation and positioning system, digital processor unit and agricultural machinery by-wire chassis system, and ensuring the stable operation of the system.

[0055] Digital processor unit: This unit is the execution core of the control program, responsible for collecting and processing relevant state variable data required in the control program, calculating and outputting control commands according to the control algorithm program of the present invention, and ensuring the precise and efficient motion control of the unmanned agricultural machinery.

[0056] Agricultural machinery by-wire chassis system: This unit is the execution mechanism of the unmanned agricultural machinery, responsible for receiving the control commands of the digital processor unit, and converting them into specific mechanical actions, managing actions such as the forward, backward and turning of the unmanned agricultural machinery, ensuring that the agricultural machinery can accurately execute according to the control commands, and realizing efficient operation.

[0057] For the embodiments of the above system, an embodiment of a dynamic obstacle avoidance method for unmanned agricultural machinery based on a guiding vector field includes the following content:

[0058] (1) In the digital processor unit, according to the motion characteristics of the driverless agricultural machinery, a kinematic model of the driverless agricultural machinery including external disturbances is established, and relevant state variables are defined: the position and heading angle of the driverless agricultural machinery. The kinematic model of the driverless agricultural machinery with external disturbance characteristics for modeling the driverless agricultural machinery is as follows:

[0059]

[0060] Among them, (x, y) represents the position coordinates of the driverless agricultural machinery in the geodetic coordinate system, ψ represents the heading angle of the driverless agricultural machinery, and both of them are obtained through the navigation and positioning system; v represents the speed of the driverless agricultural machinery, δ represents the front wheel angle of the driverless agricultural machinery, L is the wheelbase of the front and rear axles of the driverless agricultural machinery, and d 1 、d 2 、d 3 respectively represent external disturbances on different channels, including the position in the x - direction of the geodetic coordinate system, the position in the y - direction of the geodetic coordinate system, and the heading of the agricultural machinery. These disturbances reflect various interference factors that may exist in the dynamic environment, such as terrain changes, soil factors, etc.

[0061] Through accurate modeling, the kinematic model can accurately describe the motion characteristics of the driverless agricultural machinery in a complex environment, providing a solid foundation for subsequent control strategies.

[0062] (2) In the digital processor unit, a guidance vector field module is constructed, and an attractive vector field and a repulsive vector field are designed. The attractive vector field is used to guide the driverless agricultural machinery to move forward along the preset target to ensure efficient completion of the operation task, and the repulsive vector field makes the driverless agricultural machinery stay away from obstacles to avoid collisions, and the attractive vector field and the repulsive vector field are combined to form the final guidance vector field.

[0063] Construct the attractive vector field:

[0064]

[0065] Construct the repulsive vector field:

[0066]

[0067] The guidance vector field is composed of a mixture of the attractive vector field and the repulsive vector field and has the following form:

[0068]

[0069] Among them, r t =r - r target , r = (x, y) is the position of the driverless agricultural machinery, and r target is the position of the target point tracked by the driverless agricultural machinery (given in real - time by the upper computer); r s，i =r - r o，i, r o，i is the position of the i-th obstacle (obtained through the environmental perception system); F a (r i ) is the attraction vector field, F r，i (r s，i ) is the repulsive vector field of the i-th obstacle, and F(r) is the combined guidance vector field; σ i (β i ) is the mixing function, which takes values between 0 and 1; p represents an adjustable unit vector, and its value is related to the positions of the obstacle and the target point; N represents the number of obstacles; the intermediate quantity β i =-||r - r s，i || 2 , a constant β i =-||ρ b + ρ o，i + ρ a，i || 2 , a constant ρ b is the radius of the smallest circle enclosing the unmanned agricultural machinery, ρ o，i is the radius of the i-th obstacle, ρ a，i and ρ s，i are user-defined parameters. When ρ a,i and ρ s,i are larger, the range of action of the repulsive vector field is larger, and the distance between the path of the unmanned agricultural machinery and the obstacle will be farther; ξ(β i ) is a mixing function related to β i , and has the following form:

[0070]

[0071] By combining the attraction vector field and the repulsive vector field, it is ensured that the unmanned agricultural machinery can accurately track the target while avoiding obstacles, significantly enhancing the obstacle avoidance ability of the agricultural machinery in complex environments.

[0072] (3) In the digital processor unit, based on the kinematic model of the unmanned agricultural machinery and the guidance vector field, a disturbance observer is designed to monitor and accurately estimate external disturbances in real time.

[0073]

[0074] Among them, the sliding mode variable is defined as s = e ψ , the heading angle error e ψ = ψ - ψ r , ψ r represents the desired heading angle of the guidance vector field, and ψ r = arctan(F y / F x ), Fy represents the component of the guiding vector field F(r) in the y-direction, F x represents the component of the guiding vector field F(r) in the x-direction; is the estimated value of the sliding mode variable s, is the estimated value of the lumped disturbance, and are respectively the first-order derivatives of; and are artificially selected observer parameters, which can usually be selected by the bandwidth method. First, determine a positive constant ω, and then let Generally speaking, the larger ω is, the more accurate the disturbance estimation is, but it will increase the noise introduced by the output measurement. The smaller ω is, the greater the phase lag of the disturbance estimation is, resulting in inaccurate disturbance estimation; when actually selecting parameters, factors such as the bandwidth of external disturbances and the frequency of output measurement need to be comprehensively considered.

[0075] The disturbance observer can perform real-time observation and analysis on various disturbance factors that may appear in the dynamic environment, such as terrain changes, soil factors, etc., and provide accurate disturbance information for subsequent control strategies. Real-time estimation of external disturbances and feedforward compensation improve the tracking accuracy and system robustness of unmanned agricultural machinery in complex environments.

[0076] (4) Non-smooth obstacle avoidance controller design module (implemented in the digital processor unit), based on the disturbance information provided by the disturbance observer, this module is responsible for constructing a non-smooth obstacle avoidance controller with disturbance compensation function:

[0077]

[0078] where η 1 、η 2 、η 3 represent the controller parameters, all of which are positive constants, and their values are determined according to the results of field experiments of unmanned agricultural machinery. For agricultural machinery with different parameters, the values of η 1 、η 2 、η 3 are also different; α and β are positive constants, and α satisfies 0 ≤ α < 1, β satisfies 0 ≤ β ≤ 1; the sign function sign(x) is specifically expressed as:

[0079] The non-smooth obstacle avoidance controller adjusts the control strategy in real time by introducing the external disturbance estimated by the disturbance observer to compensate for the influence of the external disturbance on the motion trajectory of the unmanned agricultural machinery. At the same time, by introducing the integral term and non-smooth control technology, the tracking accuracy and anti-interference ability of the unmanned agricultural machinery in complex dynamic environments are significantly improved, ensuring that the unmanned agricultural machinery can stably and accurately track the desired heading in complex dynamic environments, and effectively enhancing the obstacle avoidance performance and operation accuracy.

[0080] (5) In the digital processor unit, design the obstacle avoidance speed adaptive strategy for the unmanned agricultural machinery: when the heading angle error of the unmanned agricultural machinery is large, the unmanned agricultural machinery will actively reduce the forward speed, so that the steer-by-wire chassis of the agricultural machinery can have enough time to execute the steering operation, avoiding potential collision risks in practical applications:

[0081]

[0082] where, v 0 is the set speed of the agricultural machinery, s th and γ are positive constants, s th is the heading angle error threshold determined by experiments, and the smaller s th is, the more sensitive the speed adaptive strategy is to the heading angle error, so that the collision risk can be reduced more effectively. The parameter γ can control the attenuation speed of the agricultural machinery speed with the heading angle error, and its value is determined by experiments.

[0083] If the parameter configuration of the above obstacle avoidance controller is improper, or the initial yaw angle error of the unmanned agricultural machinery is large, the unmanned agricultural machinery will not have enough time to steer to avoid obstacles, and there may be some collision risks. In this case, the adopted speed adaptive strategy can effectively reduce these potential collision risks.

[0084] (6) The non-smooth obstacle avoidance controller adjusts the front wheel angle (control input) of the unmanned agricultural machinery through Equation (9), sends the control command to the steer-by-wire chassis system of the agricultural machinery, and tracks the desired heading angle of the guidance vector field to achieve the purpose of dynamic obstacle avoidance.

[0085] The present invention also provides a dynamic obstacle avoidance system for unmanned agricultural machinery based on a guidance vector field, and the system includes:

[0086] An agricultural machinery kinematic model construction module: This module is responsible for establishing a kinematic model of the unmanned agricultural machinery including external disturbances and defining relevant state variables;

[0087] A guidance vector field construction module: This module focuses on designing an attractive vector field and a repulsive vector field, and combines the two to form the final guidance vector field;

[0088] A disturbance observer design module: The core task of this module is to monitor and accurately estimate external disturbances in real time;

[0089] Non-smooth obstacle avoidance controller design module: Based on the external disturbance provided by the disturbance observer, this module is responsible for constructing a non-smooth obstacle avoidance controller with disturbance compensation function;

[0090] Obstacle avoidance speed adaptive strategy module: When the heading error of the unmanned agricultural machine is large, it realizes the active speed reduction of the unmanned agricultural machine.

[0091] Examples of the method and system for obstacle avoidance of unmanned agricultural machines based on the guiding vector field have been specifically explained and illustrated, making the technical solution of the present invention clearer and easier to implement. Based on this example, through simulation field experiments, the obstacle avoidance performance and tracking performance of the present invention in complex environments have been verified. The experimental results show that in the presence of external disturbances, the system can effectively avoid obstacles and track the target.

[0092] Set the unmanned agricultural machine in a dynamic environment and test the obstacle avoidance performance of this example according to the obstacle avoidance results. The experimental results are as Figure 3 shown. In an environment with external obstacles, the control method in this example can effectively avoid obstacles and track the target. Therefore, it can be demonstrated that the present invention has excellent obstacle avoidance and tracking performance.

[0093] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0094] Although the present invention has been described according to various specific embodiments, those skilled in the art will realize that the present invention can be implemented with modifications within the spirit of the claims. Therefore, any obvious improvements, substitutions, or variations that can be made by those skilled in the art without departing from the essence of the present invention fall within the protection scope of the present invention.

Claims

1. A method for dynamic obstacle avoidance of unmanned agricultural machinery based on a guidance vector field, characterized in that: According to the motion characteristics of unmanned agricultural machinery, a kinematic model of unmanned agricultural machinery including external disturbances is established; The attraction vector field and the repulsion vector field are combined to form a guidance vector field; Based on the kinematic model of unmanned agricultural machinery and the guidance vector field, a disturbance observer is designed to monitor and accurately estimate external disturbances in real time. Based on the disturbance information provided by the disturbance observer, a non-smooth obstacle avoidance controller is designed. The non-smooth obstacle avoidance controller adjusts the front wheel steering angle of the unmanned agricultural machine and tracks the desired heading angle of the guidance vector field to achieve the purpose of dynamic obstacle avoidance. The non-smooth obstacle avoidance controller is: Among them, δ is the front wheel turning angle of the unmanned agricultural machine, L is the wheelbase of the front and rear axles of the unmanned agricultural machine, v is the speed of the unmanned agricultural machine, s is the sliding mode variable, and s = e ψ , heading error e ψ =ψ-ψ r , ψ is the heading angle of the unmanned agricultural machinery, ψ r is the desired heading angle of the guidance vector field, η1, η2, η3 are controller parameters, α and β are positive constants, is the estimated value of the lumped disturbance, and sign(x) is the sign function.

2. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 1 is characterized in that: When the heading error of the unmanned agricultural machine is greater than or equal to the set threshold, the unmanned agricultural machine will actively reduce its speed, so that the agricultural machine's wire-controlled chassis has enough time to perform steering operations, thus avoiding potential collision risks in actual applications: Among them, v0 is the set speed of the agricultural machinery, s th and γ are positive constants, s th The heading error threshold determined experimentally.

3. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 1 is characterized in that: The disturbance observer is: in, is the estimated value of the sliding mode variable s, and They are The first derivative of and is an artificially selected observer parameter.

4. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 1 is characterized in that: The guidance vector field is: Where F(r) is the guidance vector field, σ i (β i ) is the mixing function, F a (r t ) is the attraction vector field, F r,i (r s,i ) is the repulsion vector field of the ith obstacle, r t =rr target , r s，i =rr o，i ,,r is the position of the unmanned agricultural machinery, r target is the position of the target point tracked by the unmanned agricultural machinery, r o，i is the position of the ith obstacle, N is the number of obstacles, and the intermediate quantity β i =-||rr s，i || 2 ,constant β i =-||ρ b +ρ o，i +ρ a，i || 2 ,constant ρ b is the radius of the minimum circle surrounding the unmanned agricultural machinery, ρ o，i is the radius of the ith obstacle, ρ a,i and ρ s，i is a custom parameter, ξ(β i ) is related to β i The relevant mixing function, and 5. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 4 is characterized in that: The attraction vector field 6. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 4 is characterized in that: The repulsion vector field of the i-th obstacle Where p is an adjustable unit vector.

7. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 4 is characterized in that: The desired heading angle ψ of the guidance vector field r =arctan(F y / F x ), F y is the component of the guide vector field F(r) in the y direction, F x is the component of the guidance vector field F(r) in the x direction.

8. The unmanned agricultural machinery dynamic obstacle avoidance method based on the guidance vector field according to claim 1 is characterized in that: The kinematic model of the unmanned agricultural machinery including external disturbance is: Among them, d1, d2, and d3 represent external disturbances on different channels, including the position in the x direction of the geodetic coordinate system, the position in the y direction of the geodetic coordinate system, and the heading of the agricultural machinery. (x, y) represents the position coordinates of the unmanned agricultural machinery in the geodetic coordinate system.

9. A system for implementing the unmanned agricultural machinery dynamic obstacle avoidance method based on a guidance vector field as described in any one of claims 1 to 8, characterized in that: include: Agricultural machinery kinematic model building module: responsible for building the kinematic model of unmanned agricultural machinery including external disturbances; Guidance vector field construction module: design attraction vector field and repulsion vector field, and combine them to form a guidance vector field; Disturbance observer design module: Based on the kinematic model of unmanned agricultural machinery and the guidance vector field, a disturbance observer is designed to monitor and accurately estimate external disturbances in real time; Non-smooth obstacle avoidance controller design module: Based on the external disturbance provided by the disturbance observer, a non-smooth obstacle avoidance controller with disturbance compensation function is constructed; Obstacle avoidance speed adaptive strategy module: When the heading error of the unmanned agricultural machinery is greater than or equal to the set threshold, the unmanned agricultural machinery will actively reduce its speed.

10. A device for implementing the unmanned agricultural machinery dynamic obstacle avoidance method based on a guidance vector field as described in any one of claims 1 to 8, characterized in that: include: The host computer module is used to send the location of the target point tracked by the unmanned agricultural machinery; Environmental perception system to identify the location, shape and size of obstacles; Navigation and positioning system, used to continuously obtain the real-time location and path of unmanned agricultural machinery; A digital processor unit for constructing the guidance vector field, designing the disturbance observer, and the nonsmooth obstacle avoidance controller; The agricultural machinery wire-controlled chassis system receives control instructions from the digital processor unit and controls the actions of the unmanned agricultural machinery; The vehicle communication system is responsible for data exchange, ensuring data synchronization and collaborative work between the host computer module, environmental perception system, navigation and positioning system, digital processor unit and agricultural machinery wire-controlled chassis system.

Images