WO2024031959A1

WO2024031959A1 - Obstacle avoidance robot, control method and control device thereof, and readable storage medium

Info

Publication number: WO2024031959A1
Application number: PCT/CN2023/077991
Authority: WO
Inventors: 曹开发; 刘冬
Original assignee: 美的集团(上海)有限公司; 美的集团股份有限公司
Priority date: 2022-08-10
Filing date: 2023-02-23
Publication date: 2024-02-15
Also published as: CN115268457A

Abstract

An obstacle avoidance robot, a control method and control device thereof, a readable storage medium, and a computer program product. The control method comprises: receiving map information of an environment where an obstacle avoidance robot is located (S102); acquiring an initial path of the obstacle avoidance robot and an initial boundary of the initial path according to the map information (S104); detecting obstacle information in the initial boundary, determining a target path of the obstacle avoidance robot according to the initial boundary, the initial path, and the obstacle information, and generating a control instruction (S106); and according to the control instruction, controlling the obstacle avoidance robot to move (S108).

Description

Obstacle avoidance robot and control method, control device and readable storage medium thereof

This application is required to be submitted to the State Intellectual Property Office of China on August 10, 2022, with the application number "202210957649.7" and the application name of the Chinese patent application "Obstacle Avoidance Robot and Its Control Method, Control Device and Readable Storage Medium" priority, the entire contents of which are incorporated into this application by reference.

Technical field

The present application belongs to the field of obstacle avoidance robots. Specifically, it relates to an obstacle avoidance robot and its control method, control device, readable storage medium and computer program product.

Background technique

In order to avoid collisions, current obstacle avoidance robots usually add an expansion radius to obstacles, but this does not allow the obstacle avoidance robot to start avoiding obstacles from a relatively long distance, and often causes danger due to insufficient reaction time.

Application content

This application aims to solve one of the technical problems existing in the existing technology or related technology.

To this end, the first aspect of this application proposes a control method for an obstacle avoidance robot.

The second aspect of this application proposes a control device for an obstacle avoidance robot.

The third aspect of this application proposes a control device for an obstacle avoidance robot.

A fourth aspect of the application proposes a readable storage medium.

The fifth aspect of this application proposes an obstacle avoidance robot.

A sixth aspect of the application proposes a computer program product.

In view of this, according to the first aspect of the present application, a control method for an obstacle avoidance robot is proposed. The control method includes: receiving map information of the environment where the obstacle avoidance robot is located; and obtaining the initial trajectory and initial trajectory of the obstacle avoidance robot based on the map information. Initial boundary of the trajectory; detect obstacle information within the initial boundary, determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information, and generate control instructions; control the movement of the obstacle avoidance robot according to the control instructions.

According to the control method of the obstacle avoidance robot provided by this application, the map information of the environment where the obstacle avoidance robot is located is received, the obstacle information within the initial boundary is detected, and the target trajectory of the obstacle avoidance robot is determined based on the initial boundary, initial trajectory and obstacle information. , and generates control instructions, controls the obstacle avoidance robot to move according to the control instructions, and can control the obstacle avoidance robot to avoid obstacles at a long distance, thereby reducing the possibility of collision between the obstacle avoidance robot and obstacles, and reserving a certain amount of time in advance. Time allows the obstacle avoidance robot to respond promptly to unexpected situations and avoid risks, making the obstacle avoidance robot's movement trajectory smoother.

The second aspect of the present application provides a control device for an obstacle avoidance robot. The obstacle avoidance robot includes a housing, a detection component and a moving component. The control device includes a detection unit and a moving unit. The detection unit is used to control the detection component to detect objects outside the housing. Obstacle information; the mobile unit is used to control the moving parts to control the movement of the obstacle avoidance robot.

The third aspect of the present application provides a control device for an obstacle avoidance robot. The control device includes a memory and a processor. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, any of the above are implemented. The steps of the control method of the obstacle avoidance robot of a technical solution therefore have all the beneficial technical effects of the control method of the obstacle avoidance robot of any of the above technical solutions.

The fourth aspect of the present application provides a readable storage medium on which a program or instructions are stored. When the program or instructions are executed by a processor, the steps of the control method of the obstacle avoidance robot according to any of the above technical solutions are implemented. Therefore, it has All the beneficial technical effects of the control method of the obstacle avoidance robot of any of the above technical solutions.

The fifth aspect of the present application provides an obstacle avoidance robot, including the control device of the obstacle avoidance robot of the second aspect; or the control device of the obstacle avoidance robot of the third aspect; or a readable storage medium of the fourth aspect, thus having the above All the beneficial technical effects of the control method of the obstacle avoidance robot of any technical solution.

The sixth aspect of the present application provides a computer program product, including a computer program/instruction. When the computer program/instruction is executed by a processor, the steps of the control method of the obstacle avoidance robot of any of the above technical solutions are implemented, and thus has any of the above features. All the beneficial technical effects of the control method of the obstacle avoidance robot in one technical solution.

Additional aspects and advantages of the invention will be apparent from the description which follows, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present application will become apparent and readily understood from the description of the embodiments in conjunction with the following drawings, in which:

Figure 1 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 2 shows a schematic diagram of the initial trajectory and initial boundaries of the obstacle avoidance robot in the embodiment shown in Figure 1;

Figure 3 shows a schematic diagram of the target trajectory and target boundaries of the obstacle avoidance robot in the embodiment shown in Figure 1;

Figure 4 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 5 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 6 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 7 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 8 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 9 shows a control schematic diagram of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 10 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 11 shows a schematic flowchart of a control method for an obstacle avoidance robot according to an embodiment of the present application;

Figure 12 shows a control schematic diagram of the obstacle avoidance model according to an embodiment of the present application;

Figure 13 shows a structural block diagram of a control device of an obstacle avoidance robot according to an embodiment of the present application.

Among them, the corresponding relationship between the reference signs and component names in Figures 1 to 13 is:

100 obstacle avoidance robot, 102 initial trajectory, 104 initial boundary, 106 target trajectory, 108 target boundary, 110 obstacles, 112 theoretical system, 114 actual system, 116 Gaussian process, 118 estimation model, 120 obstacle avoidance model, 200 obstacle avoidance robot Control device, 202 memory, 204 processor.

Detailed ways

In order to understand the above-mentioned objects, features and advantages of the present application more clearly, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, as long as there is no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

Many specific details are set forth in the following description to fully understand the present application. However, the present application can also be implemented in other ways different from those described here. Therefore, the protection scope of the present application is not limited by the specific disclosures below. Limitations of Examples.

The following describes a control method for an obstacle avoidance robot, a control device 200 for an obstacle avoidance robot, a readable storage medium, and an obstacle avoidance robot according to some embodiments of the present application with reference to FIGS. 1 to 13 .

As shown in Figure 1, one embodiment of the present application provides a control method for an obstacle avoidance robot. The control method includes:

S102, receive map information of the environment where the obstacle avoidance robot is located;

S104. According to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S106, detect obstacle information within the initial boundary, determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information, and generate control instructions;

S108: Control the obstacle avoidance robot to move according to the control instructions.

As shown in Figure 2, the map information is received, the feasible road of the obstacle avoidance robot 100 is determined based on the map information, and the road width of the feasible road is obtained, and a collision-free initial trajectory is determined based on the feasible road and the width of the feasible road in the map information. 102. It can be understood that the initial boundaries 104 are set on both sides of the initial trajectory 102, thereby determining the feasible range of the obstacle avoidance robot 100.

Specifically, as shown in Figure 3, the information of the obstacle 110 is confirmed, and whether the obstacle 110 affects the movement of the obstacle avoidance robot 100 is determined, thereby determining the target trajectory 106, and generating control instructions to control the movement of the obstacle avoidance robot 100 according to the control instructions. , compared with directly increasing the expansion radius in related technologies, the practical feasible range of the obstacle avoidance robot 100 is increased, thereby improving the adaptability of the obstacle avoidance robot 100 to multiple usage scenarios such as household, commercial, and industrial use.

Furthermore, it can be predicted in advance before a collision occurs, and risks can be avoided at a relatively long distance, thereby improving the safety of the movement of the obstacle avoidance robot 100, so that the time of the obstacle avoidance robot 100 matches the control instructions at that time, and avoids correlation In the technology, the obstacle avoidance robot has a large delay, resulting in untimely response and collision with the obstacle 110. This makes the obstacle avoidance robot 100's moving trajectory smoother and ensures the safety of the obstacle avoidance robot 100's entire obstacle avoidance process. stability and stability, improve the user experience of the obstacle avoidance robot 100, and further increase the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

As shown in Figure 4, the control method according to one embodiment of the present application includes:

S402: Receive map information of the environment where the obstacle avoidance robot is located;

S404, according to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S406, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S408, determine the target boundary based on the target trajectory and obstacle information;

S410, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S412: Determine the control instructions of the obstacle avoidance robot at the current moment based on the target trajectory, target boundary, first coordinate and second coordinate;

S414, control the obstacle avoidance robot to move according to the control instructions.

Specifically, as shown in Figure 3, the information of the obstacle 110 is confirmed, and whether the obstacle 110 affects the movement of the obstacle avoidance robot 100 is determined, thereby ensuring that there is no obstacle in the target trajectory 106 and the target boundary 108 that will hinder the movement of the obstacle avoidance robot 100. obstacles 110, and then determine the collision-free feasible range of the obstacle avoidance robot 100, and plan the movement trajectory of the obstacle avoidance robot 100. This application determines the movement trajectory and movement boundary of the obstacle avoidance robot 100 by combining the obstacle information. Compared with related Directly increasing the expansion radius in the technology increases the practical feasible range of the obstacle avoidance robot 100, thereby improving the adaptability of the obstacle avoidance robot 100 to multiple usage scenarios such as home, commercial, and industrial use.

It can be understood that the obstacle information includes the space occupied by the obstacle 110 within the initial boundary 104 , so that the target trajectory 106 and the target boundary 108 can be determined according to the space occupied by the obstacle 110 , the initial boundary 104 and the initial trajectory 102 .

Further, the space occupied by the obstacle 110 may be displayed as a circle, rectangle, triangle or other shape on the initial boundary 104. According to the shape formed by the space occupied by the obstacle 110, it is similar to one of the two initial boundaries 104. 104 and the shortest distance between the endpoints of the shape formed by the space occupied by the obstacle 110 and the adjacent initial boundary 104 form the target boundary 108 .

Furthermore, based on the first coordinate and the second coordinate, it is determined whether the movement trajectory of the obstacle avoidance robot 100 will be interfered by the obstacle 110, thereby avoiding the occurrence of collisions between the two, and thus enabling the collision to be carried out in advance before the collision occurs. Predictably, risk avoidance begins at a relatively long distance, thereby improving the safety of movement, so that the time of the obstacle avoidance robot 100 matches the control instructions at that time, and avoids untimely response caused by the large delay of the obstacle avoidance robot in related technologies. The phenomenon of collision with the obstacle 110 occurs, and at the same time, the obstacle avoidance robot 100 starts to avoid the obstacle 110 at a longer distance, thereby reducing the possibility of encountering the obstacle 110 during the obstacle avoidance process, and making the obstacle avoidance robot 100 The movement trajectory is smoother, thereby avoiding the phenomenon that the obstacle avoidance robot 100 cannot avoid the obstacle 110 at a long distance. When encountering the obstacle 110 at a close distance, the obstacle avoidance robot 100 is affected by inertia and collides with the obstacle 110 during an emergency stop. , thereby ensuring the safety and stability of the entire obstacle avoidance process of the obstacle avoidance robot 100, improving the user experience of the obstacle avoidance robot 100, and further increasing the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories. .

Moreover, the gradient descent method is used to smooth the initial trajectory 102, so that the movement trajectory of the obstacle avoidance robot 100 can be smoother, so that the obstacle avoidance robot 100 moves along a smooth movement trajectory, and the entire obstacle avoidance process can be more complete and smoother. Improve the user experience of obstacle avoidance robot 100.

As shown in Figure 5, the control method according to one embodiment of the present application includes:

S502: Receive map information of the environment where the obstacle avoidance robot is located;

S504, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory based on the map information;

S506, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S508, determine the target boundary based on the target trajectory and obstacle information;

S510, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S512: Determine the control instructions of the obstacle avoidance robot at the current moment based on the target trajectory, target boundary, first coordinate, second coordinate and obstacle avoidance model;

S514, control the obstacle avoidance robot to move according to the control instructions.

In one embodiment of the present application, the obstacle avoidance robot is jointly determined based on the risk-avoided target trajectory 106 and target boundary 108 as well as the immediately acquired first coordinates, the immediately acquired second coordinates and the pre-set obstacle avoidance model 120 100 corresponding control instructions at the current moment, so that the obstacle avoidance model 120 can be used to improve the precise control of the obstacle avoidance robot 100, and the control problem of the obstacle avoidance robot 100 can be transformed into an optimization problem of the obstacle avoidance model 120, so that the obstacle avoidance model 120 can be used in advance. The optimization of the obstacle avoidance model 120 determines whether the obstacle avoidance robot 100 can be implemented.

Moreover, by optimizing the control of the obstacle avoidance model 120, it is determined in advance whether the target trajectory 106 is feasible, thereby controlling the obstacle avoidance robot 100 to avoid obstacles. On the one hand, it saves efficiency and facilitates the control of the obstacle avoidance robot 100. On the other hand, it can avoid The obstacle avoidance robot 100 makes unnecessary movements to improve the user experience of the obstacle avoidance robot 100, thereby increasing the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

In one embodiment of the present application, the obstacle avoidance model 120 is specifically a mathematical model that transforms the motion process of the obstacle avoidance robot 100. Specifically, the obstacle avoidance model 120 can be expressed by the following two equations:
x(k+1)=f _normal (x _k ,u _k );
x(k+1)＝f _true (x _k ,u _k );

Among them, f _normal (x _k , u _k ) represents the modeling state equation of the obstacle avoidance model 120, x represents the state quantity, x _k represents the state quantity of the obstacle avoidance model 120 at time k, and u _k represents the obstacle avoidance model at time k. The output of 120, f _true (x _k , _uk ) represents the data collection state equation of the obstacle avoidance model.

Specifically, f _normal (x _k , u _k ) represents the theoretical relationship between the state quantity of the obstacle avoidance model 120 and the output quantity of the obstacle avoidance model 120 . This theoretical relationship can be based on the relationship between the state quantity and the output quantity of the obstacle avoidance model 120 The historical relational expression and the relational expression between the state quantity and the output quantity used in the obstacle avoidance model 120 in the prior art are determined.

f _true (x _k , u _k ) represents the actual relationship between the state quantity of the obstacle avoidance model 120 and the output quantity of the obstacle avoidance model 120 . The actual relationship equation can be calculated based on the actual state quantity and the actual output quantity of the obstacle avoidance model 120 . determined after variable analysis.

Further, in the obstacle avoidance model 120 , the state quantity of the obstacle avoidance model 120 may be the coordinates of the obstacle avoidance robot 100 , and the output quantity of the obstacle avoidance model 120 may be the speed or angular velocity of the obstacle avoidance robot 100 .

In the state equation in this embodiment, there is a linear correspondence between the state quantity at time k+1 and the state quantity and input quantity at time k. Therefore, k+1 can be determined based on the state quantity and input quantity obtained at time k. The state quantity at the time is thus obtained in advance according to the obstacle avoidance model 120, and then the obstacle avoidance robot 100 is controlled.

Furthermore, there is a certain error between the modeling state equation of the obstacle avoidance model 120 and the data collection state equation of the obstacle avoidance model 120. This error is caused by modeling errors, losses in the data collection process, and model noise. The modeling state equation of the obstacle avoidance model 120 can be obtained through Gaussian compensation and the data acquisition state equation of the obstacle avoidance model can be expressed as:
x(k+1)＝f _normal (x _k ,u _k )+Δ

Among them, △~N(u,Σ), △ represents the compensation value of the obstacle avoidance model 120, u is the mean value of the input amount collected in advance by the obstacle avoidance model 120, and Σ is the input of the pre-fitting training in the obstacle avoidance model 120 quantity covariance.

In one embodiment of the present application, the obstacle avoidance model 120 is specifically a mathematical model of the motion process transformation of the obstacle avoidance robot 100. Specifically, it can be expressed as:

st

in, Expressed as the cost function equation of the obstacle avoidance model 120, st represents subject to (subject to), △ represents the compensation value of the obstacle avoidance model 120, x _t represents the state quantity of the obstacle avoidance model 120 at time t, x _start is The state quantity of the obstacle avoidance model 120 at the beginning time, x _t+N represents the state quantity of the obstacle avoidance model 120 at time t+N, x _end is the state quantity at the end time of the obstacle avoidance model 120, h(x _{t+ k} ) is expressed as the control obstacle constraint function of the obstacle avoidance model 120, expressed as:
h＝(x _state -x _obstacle ) ² + (y _state -y _obstacle ) ² -R ²

Among them, x _state represents the abscissa coordinate of the obstacle avoidance robot 100 within the moving boundary, x _obstacle represents the abscissa coordinate of the obstacle 110 within the moving boundary, y _state represents the ordinate coordinate of the obstacle avoidance robot 100 within the moving boundary, y _Obstacle represents the ordinate of the obstacle 110 within the movement boundary, and R represents the distance between the obstacle avoidance robot 100 and the obstacle 110 before the obstacle avoidance robot 100 moves.

Further, x _t+k represents the state quantity of the obstacle avoidance model 120 at time t+k, u _t+k represents the input quantity of the obstacle avoidance model 120 at time t+k, and x represents the obstacle avoidance model 120 at time t+k. The state quantity of the model 120, u represents the input quantity of the obstacle avoidance model 120. The control obstacle constraint function can be represented by the linear differential function h(x _t+k ). After h(x _t+k ) is differentiated, the differential function Δh(x _{t +k} ,u _t+k ), in the range of 0 to 1, there is always a constant r, so that the control obstacle constraint function and differential function satisfy Δh(x _t+k ,u _t+k )≥-rh(x _{t +k} ), when r is close to 0, it means that the obstacle avoidance robot 100 is far away from the obstacle 110, when r is close to 1, it means that the obstacle avoidance robot 100 is away from the obstacle 110. The obstacle robot 100 approaches the obstacle 110.

go a step further, J in represents the total cost function of the obstacle avoidance model 120, minp(x _t+N ) represents the cost function of the state quantity at time t+N, Represents the sum of the cost functions of all state quantities and input quantities before time t+N.

As shown in Figure 6, the control method according to one embodiment of the present application includes:

S602: Receive map information of the environment where the obstacle avoidance robot is located;

S604. According to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S606, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S608, determine the target boundary based on the target trajectory and obstacle information;

S610, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S612, determine the distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

S614, determine the first constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S616, determine whether the first constraint value satisfies the first control output condition. If yes, execute S618. If not, execute S604;

S618: Control the obstacle avoidance robot to move according to the control instructions.

In this embodiment, the distance compensation value is used to compensate the first constraint value, thereby improving the accuracy of the obstacle avoidance model 120 and thereby improving the accuracy of controlling the obstacle avoidance robot 100 .

Specifically, the distance compensation value can be expressed as R, and R represents the distance maintained between the obstacle avoidance robot 100 and the obstacle 110. It can be understood that R is a fixed value, which is the obstacle avoidance robot 100 before the obstacle avoidance robot 100 outputs the control command. The distance maintained from the obstacle 110 is used to compensate the obstacle avoidance model 120 with a fixed value, thereby improving the accuracy of the obstacle avoidance model 120 and thereby improving the accuracy of controlling the obstacle avoidance robot 100 .

Further, when the first constraint value satisfies the first output control condition, it is determined that the obstacle avoidance model 120 is feasible. According to the control instructions corresponding to the obstacle avoidance model 120 and the obstacle avoidance robot 100 at the current moment, the delay of the obstacle avoidance model 120 is solved. problem, to avoid the phenomenon in related technologies that the obstacle avoidance robot 100 has a large delay, resulting in untimely response and collision with the obstacle 110, thereby making the obstacle avoidance process of the entire obstacle avoidance robot 100 smoother and improving the performance of the obstacle avoidance robot 100. The user experience increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

Furthermore, the first constraint value is a constant value in the obstacle avoidance model 120 that satisfies an inequality relationship based on the control obstacle constraint function and the differential function obtained by the control obstacle constraint function.

Based on the first constraint value satisfying the first control output condition, determining the control instructions of the obstacle avoidance robot 100 at the current moment according to the obstacle avoidance model 120 includes: setting the upper limit threshold and the lower limit threshold of the first control output condition, specifically, setting the upper limit threshold is 1, set the lower threshold to 0.

Specifically, when the first constraint value is in the range of greater than the lower limit threshold and less than or equal to the upper threshold, it is determined that the first constraint value satisfies the first control output condition. When the first constraint value is not in the range of greater than the lower threshold and less than or equal to the upper threshold. When within the range, it means that the obstacle avoidance model 120 is invalid for the obstacle constraints of the obstacle avoidance robot 100, so the initial trajectory 102 and the initial boundary 104 of the obstacle avoidance robot 100 are re-determined, and the movement trajectory of the obstacle avoidance robot 100 is re-planned.

As shown in Figure 7, the control method according to one embodiment of the present application includes:

S702: Receive map information of the environment where the obstacle avoidance robot is located;

S704, according to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S706, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S708, obtain the model compensation value of the obstacle avoidance model;

S710, determine the obstacle avoidance model based on the model compensation value and the theoretical obstacle avoidance model;

S712, determine the target boundary based on the target trajectory and obstacle information;

S714, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S716, determine the distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

S718, determine the first constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S720, determine whether the first constraint value satisfies the first control output condition. If yes, execute S722. If not, execute S704;

S722: Control the obstacle avoidance robot to move according to the control instructions.

In this embodiment, the model compensation value is obtained, and the theoretical obstacle avoidance model is compensated according to the model compensation value, thereby increasing the obstacle avoidance The degree of matching between the model 120 and the actual obstacle avoidance model reduces the impact of errors, losses, noise and other factors of the obstacle avoidance model 120 on the obstacle avoidance model 120 and the impact of model uncertainty, complex nonlinearity and other factors, and improves The accuracy and robustness of the obstacle avoidance model 120 improves the accuracy of controlling the obstacle avoidance robot 100, improves the user experience of the obstacle avoidance robot 100, and increases the suitability of the obstacle avoidance robot 100 for various usage scenarios such as homes, shopping malls, and factories. The degree of matching.

Specifically, the model compensation value is the result of compensation according to the Gaussian process 116. The Gaussian process 116 is a combination of random variables in the obstacle avoidance model 120 within the exponential set, which is determined by the mathematical expectation and covariance function, and can be expressed as:
△～N(u,∑);

Among them, △ can be expressed as the model compensation value, u is the mathematical expectation, specifically in the obstacle avoidance model 120, it represents the mean value of the input quantity collected in advance, and Σ is the covariance, specifically in the obstacle avoidance model 120 it represents the pre-fitting training. The covariance of the input quantity. The theoretical obstacle avoidance model is a pre-constructed mathematical model before the obstacle avoidance robot 100 moves. It can be represented by the equation of the theoretical system 112, specifically:
x(k+1)=f _normal (x _k ,u _k );

Among them, k represents the moment at time k, k+1 represents the moment at time k+1, x _k represents the state quantity of the obstacle avoidance model 120 at time k, u _k represents the output quantity of the obstacle avoidance model 120 at time k, f _normal represents the theoretical relationship between the state quantity of the obstacle avoidance model 120 and the output quantity of the obstacle avoidance model 120 , and x represents the state quantity of the obstacle avoidance model 120 .

The actual obstacle avoidance model is a data model constructed based on the data collected during the actual movement of the obstacle avoidance robot 100, and can be represented by the equation of the actual system 114:
x(k+1)＝f _true (x _k ,u _k );

Among them, x _k represents the state quantity of the obstacle avoidance model 120 at time k, u _k represents the output quantity of the obstacle avoidance model 120 at time k, and f _true represents the actual difference between the state quantity of the obstacle avoidance model 120 and the output quantity of the obstacle avoidance model 120 In the relational expression, x represents the state quantity of the obstacle avoidance model 120.

Due to modeling errors, losses, noise and other factors of the obstacle avoidance model 120, there is a certain deviation between the actual obstacle avoidance model and the theoretical obstacle avoidance model. Therefore, the Gaussian process 116 is introduced, so that the actual obstacle avoidance model can be composed of the theoretical obstacle avoidance model and the Gaussian process. 116 is obtained, thereby compensating for the impact of model uncertainty, complex nonlinear factors, etc. on the obstacle avoidance model 120, and improving the accuracy and robustness of the obstacle avoidance model 120, which is specifically expressed as:
x(k+1)＝f _normal (x _k ,u _k )+Δ

Among them, k represents the moment at time k, k+1 represents the moment at time k+1, x _k represents the state quantity of the obstacle avoidance model 120 at time k, u _k represents the output quantity of the obstacle avoidance model 120 at time k, f _normal represents the theoretical relationship between the state quantity of the obstacle avoidance model 120 and the output quantity of the obstacle avoidance model 120 , Δ represents the model compensation value, and x represents the state quantity of the obstacle avoidance model 120 .

Further, the model compensation value can be changed periodically, and the change period can be set according to the loss of the obstacle avoidance model 120. It is understandable that the algorithm of the obstacle avoidance model 120 in the related art uses a linearized model and ignores the uncertainty of the model. factors, so the process of establishing a complex nonlinear model is very difficult, and during use of the obstacle avoidance robot 100, many parameters change due to usage losses, such as tire friction coefficient, etc., and the Gaussian process 116 is introduced to compensate for the obstacle avoidance model. The uncertainty of 120 makes the algorithm accuracy and robustness of the obstacle avoidance model 120 higher.

As shown in Figure 8, the control method according to one embodiment of the present application includes:

S802: Receive map information of the environment where the obstacle avoidance robot is located;

S804, according to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S806, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S808, obtain the model compensation value of the obstacle avoidance model;

S810, determine the obstacle avoidance model based on the model compensation value and the theoretical obstacle avoidance model;

S812, determine the target boundary based on the target trajectory and obstacle information;

S814, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S816, determine the distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

S818, determine the second constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S820, perform data processing on the second constraint value to obtain the third constraint value;

S822, determine the first constraint value based on the second constraint value and the third constraint value;

S824, determine whether the first constraint value satisfies the first control output condition. If yes, execute S826. If not, execute S804;

S826: Control the obstacle avoidance robot to move according to the control instructions.

In this embodiment, the second constraint value represents the result obtained by controlling the obstacle constraint function of the obstacle avoidance model 120. The second constraint value h can be expressed as:
h＝(x _state -x _obstacle ) ² + (y _state -y _obstacle ) ² -R ²

Among them, x _state represents the center abscissa of the obstacle avoidance robot 100 at the current moment, x _obstacle represents the center abscissa of the obstacle 110 at the current moment, y _state represents the center abscissa of the obstacle avoidance robot 100 at the current moment, and y _obstacle represents The center abscissa of the obstacle 110 at the current moment, R represents the distance between the obstacle avoidance robot 100 and the obstacle 110 before the obstacle avoidance robot 100 moves.

Further, the third constraint value is the result of differentiating the obstacle constraint function of the obstacle avoidance model 120, expressed as Δh, and the first constraint value is a constant representing the convergence speed in the obstacle avoidance model 120, which can be expressed as r, at 0 Within the range of 1, there is always a constant in the obstacle avoidance model 120, so that the first constraint value, the second constraint value and the third constraint value satisfy the following relationship:
Δh(x _t+k ,u _t+k )+rh(x _t+k )≥0;

Through the relational expression of the first constraint value, the second constraint value and the third constraint value, the inequality constraint of the obstacle avoidance model 120 by the control obstacle constraint function is realized, thereby increasing the feasible set range of the obstacle avoidance robot 100, where, h(x _{t +k} ) represents the second constraint value at the t+kth time, and Δh(x _t+k , u _t+k ) is the third constraint value obtained by differentiating the second constraint value at the t+kth time.

It can be understood that when the third constraint value approaches 0, it means that the obstacle avoidance robot 100 is far away from the obstacle 110; when the third constraint value approaches 1, it means that the obstacle avoidance robot 100 is closer to the obstacle 110. .

Further, through the inequality constraints between the first constraint value, the second constraint value and the third constraint value, the accuracy of the obstacle avoidance model 120 is improved, and the feasible set range of the obstacle avoidance model 120 is increased, so that the obstacle avoidance robot can 100 can start to avoid obstacles 110 at a longer distance, reducing the possibility of encountering obstacles 110 during the obstacle avoidance process, thereby preventing the obstacle avoidance robot 100 from being unable to avoid obstacles 110 at a long distance. When encountering obstacles 110 at close range, During an emergency stop, the obstacle avoidance robot 100 is affected by inertia and collides with the obstacle 110. This ensures the safety and stability of the entire obstacle avoidance process of the obstacle avoidance robot 100, improves the user experience of the obstacle avoidance robot 100, and further improves the user experience of the obstacle avoidance robot 100. Increase the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

As shown in Figure 9, the control method according to one embodiment of the present application includes:

S902: Receive map information of the environment where the obstacle avoidance robot is located;

S904, according to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S906, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S908, obtain the model compensation value of the obstacle avoidance model;

S910, determine the obstacle avoidance model based on the model compensation value and the theoretical obstacle avoidance model;

S912, determine the target boundary based on the target trajectory and obstacle information;

S914, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S916, determine the distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

S918, determine the second constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S920, perform data processing on the second constraint value to obtain the third constraint value;

S922, determine the first constraint value based on the second constraint value and the third constraint value;

S924, determine whether the first constraint value satisfies the first control output condition. If yes, execute S926. If not, execute S904;

S926, determine the control instructions of the obstacle avoidance robot at the current moment;

S928: Determine whether the control instruction of the obstacle avoidance robot at the current moment meets the second control output condition. If yes, execute S930; if not, execute S904;

S930: Use the control instruction of the obstacle avoidance robot at the current moment as an optimization control instruction, and control the movement of the obstacle avoidance robot according to the optimization control instruction.

In this embodiment, based on the fact that the control instruction of the obstacle avoidance robot 100 at the current moment satisfies the second control output condition, it is determined that the control instruction of the obstacle avoidance robot 100 at the current moment has been optimized, and the control instruction at the current moment is used as the optimized control instruction.

Further, controlling the movement of the obstacle avoidance robot 100 according to the optimized control instructions includes: extracting the movement parameters in the control instructions, Perform data processing on movement parameters. Specifically, the movement parameters include the speed of the obstacle avoidance robot 100, the acceleration of the obstacle avoidance robot 100, the displacement of the obstacle avoidance robot 100, the position error of the obstacle avoidance robot 100, the speed of the obstacle 110, the acceleration of the obstacle 110, the obstacle The displacement of 110, the position error of obstacle 110, etc.

Further, determine the speed of the obstacle avoidance robot 100 , the acceleration of the obstacle avoidance robot 100 , the displacement of the obstacle avoidance robot 100 , the position error of the obstacle avoidance robot 100 , the speed of the obstacle 110 , the acceleration of the obstacle 110 , and the displacement of the obstacle 110 , whether the position error of the obstacle 110 has a relative minimum value in the obstacle avoidance model 120. When the movement parameter has a relative minimum value in the obstacle avoidance model 120, it is determined that the movement trajectory of the obstacle avoidance robot 100 is the optimal one at the current moment. trajectory, thereby outputting control instructions to control the obstacle avoidance robot 100 to move according to the movement trajectory, avoiding unnecessary movements of the obstacle avoidance robot 100 and ensuring that the obstacle avoidance robot 100 moves the minimum distance, thereby improving the intelligence of the obstacle avoidance robot 100 and energy saving, improve the use experience of the obstacle avoidance robot 100, and increase the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

Specifically, the cost function relationship in the obstacle avoidance model 120 can be used to perform data processing on the movement parameters, which can be expressed as:

Among them, p(x _t+N ) represents the movement parameter at time t+N, represents the sum of all movement parameters from time t=0 to time t+N-1, and J represents the sum of all movement parameters from time t=0 to time t+N. It can be understood that the time t=0 is the starting time of the entire movement process, and t+N can be expressed as the ending time of the entire movement process.

Further, determine whether p(x _t+N ) has a minimum value, thereby determining whether the cost function in the obstacle avoidance model 120 has a minimum value. When the cost function is the minimum, the movement parameters are the optimal values, and the obstacle avoidance robot is controlled to move the minimum Movement amount to achieve optimal control of the obstacle avoidance robot 100, avoid unnecessary movements of the obstacle avoidance robot 100, improve the control efficiency of the obstacle avoidance robot 100, improve the intelligence and energy saving of the obstacle avoidance robot 100, and improve obstacle avoidance The user experience of the robot 100 increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

It can be understood that the second control output condition is to set a minimum value output instruction in the optimizer. The above functional equation relationship is obtained by solving the optimizer. When there is a minimum value in the cost function relationship, the optimizer outputs the minimum value of the cost function. , determine the movement parameters at the current moment as the minimum movement parameters, determine the control instructions at the current moment as the optimization control instructions, and control the obstacle avoidance robot to move according to the optimization control instructions.

When there is no minimum value in the cost function relationship, the control instruction at the current moment is not an optimized control instruction, and the obstacle avoidance robot 100 cannot avoid obstacles with the minimum movement parameters, so the obstacle avoidance robot 100 can re-plan its movement trajectory.

As shown in Figure 10, a control method provided by an embodiment of the present application includes:

S1002, receive map information of the environment where the obstacle avoidance robot is located;

S1004. According to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S1006, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S1008, obtain the model compensation value of the obstacle avoidance model;

S1010, determine the obstacle avoidance model based on the model compensation value and the theoretical obstacle avoidance model;

S1012, determine the target boundary based on the target trajectory and obstacle information;

S1014, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S1016, determine the distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

S1018, determine the second constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S1020, perform data processing on the second constraint value to obtain the third constraint value;

S1022, determine the first constraint value based on the second constraint value and the third constraint value;

S1024, determine whether the first constraint value satisfies the first control output condition. If yes, execute S1026. If not, execute S1004;

S1026, determine the control instructions of the obstacle avoidance robot at the current moment;

S1028, determine whether the control instruction of the obstacle avoidance robot at the current moment meets the second control output condition. If yes, execute S1030; if not, execute S1004;

S1030, use the control instructions of the obstacle avoidance robot at the current moment as the optimization control instructions, and control the movement of the obstacle avoidance robot according to the optimization control instructions;

S1032: Set the time when the obstacle avoidance robot stops moving according to the optimization control instruction as the first time, and obtain the fourth coordinates of the obstacle avoidance robot at the first time.

In this embodiment, after the obstacle avoidance robot 100 stops, the time when it stops moving is regarded as the first time, and the optimization control instruction is Corresponding to the time, the delay problem of the obstacle avoidance robot 100 can be solved, and the phenomenon in related technologies that the obstacle avoidance robot 100 has a large delay resulting in untimely response and collision with obstacles can be avoided, thereby making the obstacle avoidance robot 100 more efficient. The obstacle safety performance is higher, the user experience of the obstacle avoidance robot 100 is improved, and the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories is increased.

Specific examples:

As shown in Figure 11, the control methods provided by this application include:

S1102, receive map information of the environment where the obstacle avoidance robot is located;

S1104. According to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

S1106, smooth the initial trajectory;

S1108, detect obstacle information within the initial boundary, and determine the target trajectory of the obstacle avoidance robot based on the initial boundary, initial trajectory and obstacle information;

S1110, obtain the model compensation value of the obstacle avoidance model;

S1112, determine the obstacle avoidance model based on the model compensation value and the theoretical obstacle avoidance model;

S1114, determine the target boundary based on the target trajectory and obstacle information;

S1116, obtain the first coordinate of the obstacle and the second coordinate of the obstacle avoidance robot;

S1118, determine the distance compensation value based on the third coordinate and the second coordinate of the starting point of the target trajectory;

S1120, determine the second constraint value based on the first coordinate, the second coordinate and the distance compensation value;

S1122, perform data processing on the second constraint value to obtain the third constraint value;

S1124, determine the first constraint value based on the second constraint value and the third constraint value;

S1126, pre-set the output range of the first constraint value;

S1128, determine whether the first constraint value is within the output range, if so, execute S1130, if not, execute S1104;

S1130, determine that the first constraint value satisfies the first control output condition, and determine the control instructions of the obstacle avoidance robot at the current moment according to the obstacle avoidance model;

S1132, extract the movement parameters in the control instruction and perform data processing on the movement parameters;

S1134, determine whether the movement parameters after data processing have a minimum solution in the obstacle avoidance model. If so, execute S1136; if not, execute S1104;

S1136, determine that the control command at the current moment meets the second control output condition, use the control command of the obstacle avoidance robot at the current moment as the optimized control command, and control the movement of the obstacle avoidance robot according to the optimized control command;

S1138, set the moment when the obstacle avoidance robot stops moving according to the optimization control instruction as the first moment, and obtain the fourth coordinate of the obstacle avoidance robot at the first moment;

S1140, determine whether the fourth coordinate is the same as the fifth coordinate of the end point of the target trajectory. If yes, stop; if not, execute S1142;

S1142, obtain the obstacle information at the first moment;

S1144, determine the final trajectory of the obstacle avoidance robot based on the target boundary, target trajectory and obstacle information at the first moment;

S1146, determine the final boundary based on the final trajectory and obstacle information;

S1148, obtain the sixth coordinate of the obstacle at the first moment and the seventh coordinate of the obstacle avoidance robot at the first moment;

S1150: Determine the optimal control instruction of the obstacle avoidance robot at the first moment based on the final trajectory, the final boundary, the sixth coordinate and the seventh coordinate, until the obstacle avoidance robot is controlled to reach the eighth coordinate of the end point of the final trajectory according to the optimization control instruction.

In one embodiment of the present application, the A*(A-Star) algorithm may be used to obtain the initial trajectory 102. The A*(A-Star) algorithm may be expressed as:
F(n)=G(n)+H(n)

Among them, the A* algorithm sets multiple grid points between the initial position of the obstacle avoidance robot 100 and the target position of the obstacle avoidance robot 100, thereby converting the distance problem between the initial position and the target position into multiple points. The conversion problem improves the accuracy of controlling the obstacle avoidance robot 100 during its movement from the initial position to the target position.

Among them, F(n) represents the cost relationship from the initial position of the obstacle avoidance robot 100 to the target position, and G(n) represents the cost relationship from the initial position to the next grid point in the A* algorithm; H(n) Expressed as the cost relationship from the next grid point to the target position in the A* algorithm.

Further, the A* algorithm is used to search for feasible routes of the obstacle avoidance robot 100 and generate a list of feasible routes. The optimal trajectory is searched for in the list of feasible routes, and the optimal trajectory is used as the initial trajectory 102 .

It can be understood that the control instructions include movement parameters of the speed and angular velocity of the obstacle avoidance robot 100 .

In this embodiment, it is determined whether the fourth coordinate and the fifth coordinate are the same, thereby determining whether the obstacle avoidance robot 100 reaches the target position.

Further, if they are the same, it is determined that the obstacle avoidance robot 100 has reached the target position, thereby determining that the obstacle avoidance robot 100 has completed the complete obstacle avoidance process, and the obstacle avoidance robot 100 is controlled to stop, and the obstacle avoidance process is safe and smooth, thereby improving the performance of the obstacle avoidance robot 100. The user experience increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

Further, if different, the obstacle information is reacquired; and the final trajectory and final boundary are determined. Combined with the obstacle information, the movement trajectory and movement boundary of the obstacle avoidance robot 100 can be redetermined, so that the movement of the obstacle avoidance robot 100 can be determined based on the time. The trajectory and movement boundary are dynamically planned, so that a collision-free movement trajectory of the obstacle avoidance robot 100 can be planned from a long distance, to avoid the impact of the movement of the obstacle 110 on the obstacle avoidance robot 100, and to reduce the obstacles encountered during the obstacle avoidance process. 110 possibility, making the movement trajectory of the obstacle avoidance robot 100 smoother, and avoiding the obstacle avoidance robot 100 in related technologies that cannot avoid the obstacle 110 at a long distance. The influence of inertia causes the occurrence of dangerous collisions with obstacles 110, which improves the user experience of the obstacle avoidance robot 100 and increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

Furthermore, the sixth and seventh coordinates are obtained; the optimization control instructions are redetermined, and the obstacle avoidance robot is controlled to move by determining and outputting the optimization control instructions corresponding to the time, thus solving the delay control problem of the obstacle avoidance robot until The eighth coordinate of the obstacle avoidance robot 100 ensures the safety and stability of the entire obstacle avoidance process of the obstacle avoidance robot 100, improves the user experience of the obstacle avoidance robot 100, and further increases the impact of the obstacle avoidance robot 100 on families, shopping malls, factories, etc. The degree of adaptability to various usage scenarios.

In this embodiment, the initial trajectory 102 is obtained through the A* algorithm, and the initial trajectory 102 is smoothed to make the motion trajectory of the obstacle avoidance robot 100 smoother. Through the obstacle information, the initial boundary 104 and the smoothed initial trajectory are 102 performs dynamic planning to determine the target trajectory 106 and target boundary 108, control the movement trajectory of the obstacle avoidance robot 100 in combination with the time, and solve the obstacle avoidance model 120 according to the first control output condition and the second control output condition. , by determining whether the obstacle avoidance model 120 has a solution, and judging whether the planned movement trajectory of the obstacle avoidance robot 100 is feasible, so that the obstacle avoidance robot 100 can avoid the obstacle 110 from a long distance, and controlling the obstacle avoidance robot 100 at all times solves the problem The obstacle avoidance robot 100 is not controlled in a timely manner to avoid the delay control of the obstacle avoidance robot 100 causing the obstacle avoidance robot 100 to collide with the obstacle 110 due to the influence of inertia during emergency stop. The feasible set range of the obstacle avoidance model 120 is increased. By constraining and optimizing the obstacle avoidance model 120, the obstacle avoidance model 120 is made more accurate and robust, which facilitates the control of the obstacle avoidance robot 100 and improves the obstacle avoidance robot. 100% usage experience further increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls and factories.

Further, based on the first coordinate and the second coordinate, it is determined whether the movement trajectory of the obstacle avoidance robot 100 will be interfered by the obstacle 110, thereby avoiding the occurrence of collisions between the two, and thereby predicting risks in advance before the collision occurs. , start to avoid risks at a relatively long distance to avoid the obstacle avoidance robot 100 coming into contact with the obstacle 110 at a close distance and not responding in time, resulting in dangerous phenomena such as contact or collision between the two, affecting the safety of the obstacle avoidance robot 100. Move smoothly, thereby improving the user experience of the obstacle avoidance robot 100 and further increasing the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls, and factories.

Moreover, since the control instructions of the obstacle avoidance robot 100 correspond to the time, on the one hand, the delay problem of the obstacle avoidance model 120 is solved, which facilitates the optimal control of the obstacle avoidance model 120, and because constraints are added to the obstacle avoidance model 120 and compensation, while increasing the feasible set range of the obstacle avoidance model 120, it also improves the accuracy and robustness of the obstacle avoidance model 120, thereby improving the response speed of the obstacle avoidance model 120 and enhancing the adaptability of the obstacle avoidance model 120 and the obstacle avoidance robot 100. On the other hand, the delay problem of the obstacle avoidance robot 100 is solved, and the time corresponds to the control command. It is understandable that there is a corresponding control command at each time, so that the obstacle avoidance robot 100 can respond to the target at a remote location. Obstacles 110 are avoided, and a collision-free and smooth movement trajectory is dynamically planned, thereby making the entire movement process smoother, and improving the accuracy of controlling the obstacle avoidance robot 100, and controlling the obstacle avoidance robot 100 to start from a longer distance. Avoid obstacles 110, detect obstacle information in real time, control in advance, and adjust and control the movement of the obstacle avoidance robot 100 based on feedback, so that the obstacle avoidance robot 100 can reach the target position safely and without collision, improving the performance of the obstacle avoidance robot 100. The user experience further increases the adaptability of the obstacle avoidance robot 100 to various usage scenarios such as homes, shopping malls and factories.

Specifically, as shown in Figure 12, an actual system 114 and a theoretical system 112 are preset. The actual system 114 and the theoretical system 112 form an estimation model 118 based on the feedback value of the system. The Gaussian process 116 is in the obstacle avoidance model 120, and the estimation model is 118 is compensated. After the obstacle avoidance model 120 constrains and optimizes the estimated model 118, the actual system 114 and the theoretical system 112 are optimized and controlled.

One embodiment of the present application provides a control device 200 for an obstacle avoidance robot. The obstacle avoidance robot 100 is composed of a housing, a detection component and a moving component. The control device is composed of a detection unit and a moving unit. The detection component is controlled by the detection unit. , detecting obstacles that may hinder the movement of the obstacle avoidance robot 100; the moving parts are controlled by the mobile unit and can make the obstacle avoidance robot 100 move.

In this embodiment, the control device includes a detection unit and a moving unit. The detection unit can control the detection components of the obstacle avoidance robot 100 to detect obstacle information outside the housing of the obstacle avoidance robot 100. The mobile unit can control the moving components of the obstacle avoidance robot 100. The obstacle avoidance robot 100 is moved, so that the obstacle avoidance robot 100 can avoid the obstacle 110 and move, thereby completing the complete motion process of obstacle avoidance, improving the use experience of the obstacle avoidance robot 100, and increasing the impact of the obstacle avoidance robot 100 on household, commercial and industrial applications. The degree of adaptability to multiple usage scenarios.

Specifically, the detection component may include a sensor. This application does not limit the type of the sensor. The moving component may include a motor and a tire. This application does not specifically limit the type of the motor, the type of the tire, and the way the motor drives the tire.

As shown in Figure 13, one embodiment of the present application provides a control device 200 for an obstacle avoidance robot. The control device includes a memory 202 and a processor 204. The memory 202 stores programs or instructions, and the processor 204 can execute the stored program. or instructions to implement the steps of the control method in any of the above embodiments.

One embodiment of the present application provides a readable storage medium storing programs or instructions to implement the steps of the control method in any of the above embodiments.

One embodiment of the present application provides an obstacle avoidance robot 100, including the control device of the above embodiment; or the control device 200 of the obstacle avoidance robot of the above embodiment; or the readable storage medium of the above embodiment.

In one embodiment of the present application, the obstacle avoidance robot 100 includes a display device that can display the movement parameters, movement trajectories, movement boundaries and obstacle information in the control instructions of the obstacle avoidance robot 100 at each moment during the obstacle avoidance movement.

In this embodiment, the obstacle avoidance robot 100 also includes a display device, which displays the movement parameters, movement trajectories, movement boundaries and obstacle information in the control instructions during the movement of the obstacle avoidance robot 100, thereby increasing the user's understanding of the obstacle avoidance robot 100's avoidance. understanding of the failure process, thereby improving the user experience.

One embodiment of the present application provides an obstacle avoidance robot 100. The obstacle avoidance robot 100 includes a receiving device, a map conversion device, a detection processing device and a mobile device. The receiving device is used to receive map information of the environment where the obstacle avoidance robot 100 is located; The map conversion device is used to obtain the initial trajectory 102 of the obstacle avoidance robot 100 and the initial boundary 104 of the initial trajectory 102 based on the map information; the detection processing device is used to detect the obstacle information within the initial boundary 104, and obtain the initial trajectory 102 based on the initial boundary 104 and the initial trajectory. 102 and obstacle information, determine the target trajectory 106 of the obstacle avoidance robot 100, and generate control instructions; the mobile device is used to control the movement of the obstacle avoidance robot 100 according to the control instructions.

In this embodiment, the receiving device may have a GPS (Global Positioning System), and receive map information of the environment where the obstacle avoidance robot 100 is located at the current moment based on positioning.

The map conversion device can convert the map information of the receiving device into movement trajectories and movement boundaries. The detection and processing device can detect obstacle information, determine whether the obstacle information will affect the movement trajectory and movement boundaries, and then plan the movement trajectory and generate control instructions. The mobile device controls movement according to the control instructions until the movement reaches the end point. The receiving device, the map conversion device, the detection processing device and the mobile device interact to jointly realize the control method of any of the above embodiments, and therefore have all the beneficial technical effects of the control method of the obstacle avoidance robot 100 of any of the above embodiments.

One embodiment of the present application provides a computer program product, which includes a computer program/instruction that implements the steps of the control method of any of the above embodiments when executed by a processor. Therefore, it has all the beneficial technical effects of the control method of the obstacle avoidance robot 100 in any of the above embodiments.

The terms "first" and "second" features in the description and claims of this application may include one or more of these features, either explicitly or implicitly. In the literal description of this application, unless otherwise stated, "plurality" means two or more than two. In addition, "and/or" in the description and claims indicates at least one of the connected objects, and the character "/" generally indicates that the related objects are in an "or" relationship.

In the text description of this application, it will be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "Back", "Left", "Right", "Vertical", "Horizontal", "Top", "Bottom", "Inside", "Outside", "Clockwise", "Counterclockwise", "Axis" The orientations or positional relationships indicated by "radial direction", "circumferential direction", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the technical solutions of the present application and simplifying the description of the technical solutions of the present application. It is indicated or implied that the structures, devices, and elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore these descriptions should not be construed as limitations on the present application.

In the text description of this application, it can be understood that, unless there are clear provisions and limitations, the terms "installation", "connection" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, It can also be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be internal to two components. Connected. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood on a case-by-case basis.

In the claims, description and drawings of this application, the term "plurality" refers to two or more than two, unless there is additional explicit limitation, and the terms "upper", "lower", etc. indicate the orientation or position. The relationship is based on the orientation or positional relationship shown in the drawings, which is only for the purpose of describing the present application more conveniently and making the description process simpler, and is not intended to indicate or imply that the device or element referred to must have the specific orientation described. construction and operation in specific directions, therefore these descriptions are not to be construed as Please limit the restrictions; the terms "connection", "installation", "fixing", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection between multiple objects, or it can be a fixed connection between multiple objects. Detachable connection, or integral connection; it can be a direct connection between multiple objects, or an indirect connection between multiple objects through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood based on the specific circumstances of the above data.

In the claims, description and drawings of this application, the description of the terms "one embodiment", "some embodiments", "specific embodiments", etc. means the specific features, structures described in connection with the embodiment or example , materials or features are included in at least one embodiment or example of the present application. In the claims, description and drawings of this application, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and changes. Any modifications, equivalent substitutions, etc. made within the spirit and principles of this application shall be included in the protection scope of this application.

Claims

A control method for an obstacle avoidance robot, wherein the control method includes:

Receive map information of the environment where the obstacle avoidance robot is located;

According to the map information, obtain the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory;

Detect obstacle information within the initial boundary, determine the target trajectory of the obstacle avoidance robot based on the initial boundary, the initial trajectory and the obstacle information, and generate a control instruction;

Control the obstacle avoidance robot to move according to the control instructions.
The control method of an obstacle avoidance robot according to claim 1, wherein obstacle information within the initial boundary is detected, and the obstacle avoidance robot is determined based on the initial boundary, the initial trajectory and the obstacle information. The target trajectory and generated control instructions include:

Determine the target boundary according to the target trajectory and the obstacle information;

Obtain the first coordinates of the obstacle and the second coordinates of the obstacle avoidance robot;

According to the target trajectory, the target boundary, the first coordinate and the second coordinate, the control instruction of the obstacle avoidance robot at the current moment is determined.
The control method of the obstacle avoidance robot according to claim 2, wherein the control of the obstacle avoidance robot at the current moment is determined based on the target trajectory, the target boundary, the first coordinate and the second coordinate. Instructions include:

According to the target trajectory, the target boundary, the first coordinate, the second coordinate and the obstacle avoidance model, the control instruction of the obstacle avoidance robot at the current moment is determined.
The control method of an obstacle avoidance robot according to claim 3, wherein the avoidance method at the current moment is determined based on the target trajectory, the target boundary, the first coordinate, the second coordinate and the obstacle avoidance model. The control instructions of the obstacle robot include:

Determine a distance compensation value according to the third coordinate and the second coordinate of the starting point of the target trajectory;

Determine a first constraint value according to the first coordinate, the second coordinate and the distance compensation value;

Based on the first constraint value satisfying the first control output condition, the control instruction of the obstacle avoidance robot at the current moment is determined according to the obstacle avoidance model.
The control method of the obstacle avoidance robot according to claim 4, wherein based on the first constraint value satisfying the first control output condition, determining the control instruction of the obstacle avoidance robot at the current moment according to the obstacle avoidance model includes:

Preset the output range of the first constraint value;

Based on the first constraint value being in the output range, it is determined that the first constraint value satisfies the first control output condition, and the control instruction of the obstacle avoidance robot at the current moment is determined according to the obstacle avoidance model.
The control method of the obstacle avoidance robot according to claim 4, wherein based on the first constraint value satisfying the first control output condition, before determining the control instruction of the obstacle avoidance robot at the current moment according to the obstacle avoidance model, The control method also includes:

Obtain the model compensation value of the obstacle avoidance model;

The obstacle avoidance model is determined based on the model compensation value and the theoretical obstacle avoidance model.
The control method of an obstacle avoidance robot according to claim 4, wherein determining the first constraint value according to the first coordinate, the second coordinate and the distance compensation value includes:

Determine the second constraint value according to the first coordinate, the second coordinate and the distance compensation value;

Perform data processing on the second constraint value to obtain a third constraint value;

The first constraint value is determined based on the second constraint value and the third constraint value.
The control method of an obstacle avoidance robot according to claim 4, wherein the avoidance method at the current moment is determined based on the target trajectory, the target boundary, the first coordinate, the second coordinate and the obstacle avoidance model. The control instructions of the obstacle robot also include:

Based on the fact that the first constraint value does not satisfy the first control output condition, the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory are obtained again based on the map information.
The control method of the obstacle avoidance robot according to claim 8, wherein the control of the obstacle avoidance robot at the current moment is determined based on the target trajectory, the target boundary, the first coordinate and the second coordinate. After the instruction, the control method also includes:

Based on the fact that the control instruction of the obstacle avoidance robot at the current moment satisfies the second control output condition, the control instruction of the obstacle avoidance robot at the current moment is used as an optimized control instruction, and the movement of the obstacle avoidance robot is controlled according to the optimized control instruction.
The control method of the obstacle avoidance robot according to claim 9, wherein based on the control instruction of the obstacle avoidance robot at the current moment satisfying the second control output condition, the control instruction of the obstacle avoidance robot at the current moment is used as the optimized control Instructions, controlling the movement of the obstacle avoidance robot according to the optimization control instructions include:

Extract the movement parameters in the control instructions, and perform data processing on the movement parameters;

Based on the movement parameters after data processing, there is a minimum solution in the obstacle avoidance model, it is determined that the control instruction at the current moment satisfies the second control output condition, and the control instruction of the obstacle avoidance robot at the current moment is used as the optimized control instruction, The obstacle avoidance robot is controlled to move according to the optimization control instructions.
The control method of the obstacle avoidance robot according to claim 9, wherein based on the control instruction of the obstacle avoidance robot satisfying the second control output condition, the control instruction of the obstacle avoidance robot at the current moment is set as an optimized control instruction, After controlling the movement of the obstacle avoidance robot according to the optimized control instructions, the control method further includes:

The time when the obstacle avoidance robot stops moving according to the optimization control instruction is set as the first time, and the fourth coordinate of the obstacle avoidance robot at the first time is obtained.
The control method of the obstacle avoidance robot according to claim 11, wherein the time when the obstacle avoidance robot stops moving according to the optimization control instruction is set as the first time, and the third time of the obstacle avoidance robot at the first time is obtained. After four coordinates, the control method also includes:

Determine whether the fourth coordinate is the same as the fifth coordinate of the end point of the target trajectory;

When the fourth coordinate is the same as the fifth coordinate, control the obstacle avoidance robot to stop moving;

When the fourth coordinate is different from the fifth coordinate, obtain the obstacle information at the first moment;

Determine the final trajectory of the obstacle avoidance robot based on the target boundary, the target trajectory and the obstacle information at the first moment;

Determine the final boundary according to the final trajectory and the obstacle information; obtain the sixth coordinate of the obstacle at the first moment and the seventh coordinate of the obstacle avoidance robot at the first moment;

According to the final trajectory, the final boundary, the sixth coordinate and the seventh coordinate, the optimized control instruction of the obstacle avoidance robot at the first moment is determined until the obstacle avoidance robot is controlled according to the optimized control instruction. The robot reaches the eighth coordinate of the end point of the final trajectory.
The control method of an obstacle avoidance robot according to any one of claims 1 to 12, wherein, according to the map information, obtaining the initial trajectory of the obstacle avoidance robot and the initial boundary of the initial trajectory includes;

Based on the map information and the initial boundary, a two-dimensional coordinate grid is established.
A control device for an obstacle avoidance robot, wherein the obstacle avoidance robot includes a housing, a detection component and a moving component, and the control device includes:

A detection unit, used to control the detection component to detect obstacle information outside the housing;

A mobile unit, used to control the mobile component to control the movement of the obstacle avoidance robot.
A control device for an obstacle avoidance robot, wherein the control device includes a memory and a processor, the memory stores programs or instructions that can be run on the processor, and the program or instructions are processed by the processor. When the device is executed, the steps of the control method of the obstacle avoidance robot according to any one of claims 1 to 13 are realized.
A readable storage medium with a program or instructions stored thereon, wherein when the program or instructions are executed by a processor, the control method of the obstacle avoidance robot according to any one of claims 1 to 13 is implemented. step.
An obstacle avoidance robot, including:

The control device of an obstacle avoidance robot as claimed in claim 14; or

The control device of the obstacle avoidance robot as claimed in claim 15; or

The readable storage medium of claim 16.
A computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the steps of the control method of the obstacle avoidance robot according to any one of claims 1 to 13 are implemented.