Method, device and equipment for adjusting navigation of robot shape and size, and storage medium

Abstract

The application discloses a navigation method and device for adjusting the shape and size of a robot, an equipment and a storage medium, comprising: obtaining a navigation instruction; generating a moving route of the robot by using a navigation model according to the navigation instruction, and adjusting the shape and size of the robot. In this way, in the application, the shape and size of the robot are considered in the navigation model, and the navigation model trained by reinforcement learning generates the moving route of the robot and adaptively adjusts the shape and size of the robot after receiving the navigation instruction, so that manual adjustment of parameters is not required, and the navigation efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of robot navigation, and in particular to navigation methods, apparatus, devices, and storage media for adjusting the shape and size of robots. Background Technology

[0002] With the continuous development of autonomous mobile robots, humans are applying them to various scenarios, such as moving goods in factories or restaurants. Correspondingly, this requires autonomous mobile robots to navigate efficiently and safely in complex indoor environments.

[0003] Current navigation technologies, such as the dynamic window method, rely on local obstacle avoidance for decision-making. This requires specialized robotics engineers to design and fine-tune parameters for different robots based on their size and shape to achieve suitable navigation. However, this navigation technology is highly dependent on expert experience, leading to low efficiency.

[0004] Therefore, how to improve navigation efficiency is a key issue of concern to those skilled in the art. Summary of the Invention

[0005] To address the aforementioned problems, this application provides a navigation method, apparatus, device, and storage medium for adjusting the shape and size of a robot, thereby improving navigation efficiency. The embodiments of this application disclose the following technical solutions:

[0006] In a first aspect, this application discloses a navigation method for adjusting the shape and size of a robot, including:

[0007] Get navigation instructions;

[0008] The navigation model is used to generate the robot's path according to the navigation instructions, and the shape and size of the robot are adjusted.

[0009] Optionally, the navigation instructions include the current navigation environment, and after obtaining the navigation instructions, the system further includes:

[0010] Obtain the number of people and their trajectories corresponding to the current navigation environment;

[0011] Adjusting the shape and size of the robot includes:

[0012] The shape and size of the robot are adjusted based on the number of people and their trajectories.

[0013] Optionally, before obtaining navigation instructions, the method further includes:

[0014] Obtain multiple different navigation environments and the corresponding number of people and their trajectories under each of these different navigation environments.

[0015] Optionally, the target navigation environment can be any one of the multiple different navigation environments, the target number of personnel can be any one of the corresponding personnel numbers in the multiple different navigation environments, and the target personnel trajectory can be any one of the corresponding personnel trajectories in the multiple different navigation environments. The robot's path and shape / size differ in different navigation environments. The construction process of the navigation model is as follows:

[0016] The target navigation environment is obtained, and sample navigation labels corresponding to the target navigation environment are determined based on the number of target personnel and the trajectory of the target personnel. The sample navigation labels are used to determine the robot's path and shape size corresponding to the navigation command.

[0017] The target navigation environment is input into the initial navigation model to obtain the navigation label to be determined;

[0018] Based on the difference between the sample navigation labels and the undetermined navigation labels, the initial navigation model is adjusted to obtain the navigation model.

[0019] Optional, also includes:

[0020] The robot is controlled to move based on the path by using a reward function.

[0021] Secondly, this application discloses a navigation device for adjusting the shape and size of a robot, including:

[0022] The acquisition module is used to acquire navigation instructions;

[0023] The generation module is used to generate the robot's path using a navigation model according to the navigation instructions, and to adjust the robot's shape and size.

[0024] Optional, also includes:

[0025] The first acquisition subunit is used to acquire the number of people and their trajectories corresponding to the current navigation environment;

[0026] The generation module includes:

[0027] An adjustment module is used to adjust the shape and size of the robot based on the number of people and their trajectory.

[0028] Optional, also includes:

[0029] The second acquisition subunit is used to acquire multiple different navigation environments and the number of people and their trajectories corresponding to the multiple different navigation environments.

[0030] Optionally, the construction process of the navigation model is as follows:

[0031] The third acquisition subunit is used to acquire the target navigation environment, and determine the sample navigation label corresponding to the target navigation environment based on the number of target personnel and the trajectory of the target personnel. The sample navigation label is used to determine the robot's path and shape size corresponding to the navigation command.

[0032] The acquisition module is used to input the target navigation environment into the initial navigation model to obtain the navigation label to be determined;

[0033] An adjustment module is used to adjust the initial navigation model based on the difference between the sample navigation label and the undetermined navigation label, thereby obtaining the navigation model.

[0034] Optional, also includes:

[0035] A control module is used to control the robot to move based on the path using a reward function.

[0036] Thirdly, embodiments of this application provide an electronic device, including:

[0037] Memory, used to store computer programs;

[0038] A processor is used to implement the steps of the navigation method for adjusting the shape and size of the robot when executing the computer program.

[0039] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the navigation method for adjusting the shape and size of a robot described above.

[0040] Compared with the prior art, this application has the following beneficial effects:

[0041] This application first obtains navigation instructions, then generates the robot's path using a navigation model based on the navigation instructions, and adjusts the robot's shape and size. Thus, in this application, the robot's shape and size are taken into account in the navigation model. The navigation model, trained through reinforcement learning, automatically generates the robot's path and adaptively adjusts the robot's shape and size upon receiving navigation instructions, eliminating the need for manual parameter adjustments and improving navigation efficiency. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart of a navigation method for adjusting the shape and size of a robot provided in an embodiment of this application;

[0044] Figure 2 Simulation diagrams provided for embodiments of this application;

[0045] Figure 3 Logarithmic maps and pedestrian maps provided for embodiments of this application;

[0046] Figure 4 A schematic diagram of the structure of a navigation device for adjusting the shape and size of a robot provided in an embodiment of this application;

[0047] Figure 5 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0049] It should be noted that the navigation method, apparatus, device, and storage medium for adjusting the shape and size of a robot provided in this application are for the field of robot navigation. The above are merely examples and do not limit the application field of the methods and apparatus provided in this application.

[0050] As described earlier, with the continuous development of autonomous mobile robots, humans are applying them to various scenarios, such as moving goods in factories or restaurants. Correspondingly, this requires autonomous mobile robots to navigate efficiently and safely in complex indoor environments. Current navigation technologies rely on local obstacle avoidance methods such as the dynamic window method, which depend on carefully designed cost functions for decision-making. This requires specialized robotic engineers to design and debug parameters for different robots based on their size and shape to adjust suitable navigation parameters. However, this navigation technology is highly dependent on expert experience, leading to low navigation efficiency. Therefore, how to improve navigation efficiency is a key issue of concern for those skilled in the art.

[0051] Therefore, the inventors proposed the technical solution of this application. This application first obtains navigation instructions, then generates the robot's path using a navigation model based on the navigation instructions, and adjusts the robot's shape and size. Thus, in this application, the robot's shape and size are taken into account in the navigation model. After receiving navigation instructions, the navigation model, trained through reinforcement learning, will automatically generate the robot's path and adaptively adjust the robot's shape and size, eliminating the need for manual parameter adjustment and improving navigation efficiency.

[0052] The method provided in this application can be executed by software on a terminal device. The terminal device may be, for example, a mobile phone, tablet computer, or computer. The software may be, for example, system software.

[0053] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Method Implementation Examples

[0055] The following embodiment illustrates the navigation method for adjusting the shape and size of a robot provided in this application.

[0056] See Figure 1 The figure is a flowchart of a navigation method for adjusting the shape and size of a robot provided in an embodiment of this application. Figure 1 As shown, the method includes:

[0057] S101: Obtain navigation instructions.

[0058] In this step, the server receives navigation instructions from the client. These instructions instruct the server to control the robot to move from the starting point to the destination, and the server selects the optimal navigation route for the robot to proceed. For example, if a factory needs to move goods, it sends navigation instructions to the server, which then controls the robot to move towards the factory. The navigation instructions also include the navigation environment between the client and the robot. This environment includes the positions of multiple objects, allowing the robot to avoid obstacles such as shelves in the factory during subsequent operations.

[0059] Furthermore, before obtaining navigation commands, multiple different navigation environments and the corresponding number of personnel and their trajectories are acquired to construct a navigation model. Specifically, the target navigation environment is first acquired. Based on the target number of personnel and their trajectories, sample navigation labels are determined for the target navigation environment. These sample navigation labels are used to determine the robot's path and shape / size corresponding to the navigation commands. Then, the target navigation environment is input into the initial navigation model to obtain pending navigation labels. Finally, the initial navigation model is adjusted based on the differences between the sample navigation labels and the pending navigation labels to obtain the final navigation model. Here, the target navigation environment can be any one of the multiple different navigation environments, the target number of personnel can be any one of the corresponding number of personnel in the multiple different navigation environments, and the target personnel trajectory can be any one of the corresponding personnel trajectories in the multiple different navigation environments. It should also be noted that the corresponding robot path and shape / size are different in different navigation environments. For example, in navigation environment A, the corresponding robot path is straight and the shape / size is rectangular; in navigation environment B, the corresponding robot path is curved and the shape / size is circular. In this way, a navigation model is constructed by combining the navigation environment with the number of people and their trajectories in the navigation environment, so as to adaptively adjust the appropriate navigation strategy in different navigation environments, thereby ensuring the safety and efficiency of navigation.

[0060] In some feasible implementations, a deep reinforcement learning model (i.e., a navigation model) is trained in a 2D simulator. Specifically, a simulator for the overall navigation task is implemented using C++, and OpenCV is used for simulator visualization. The Social Force Model algorithm is used as the control strategy for regular pedestrians. See also Figure 2 This figure is a simulation diagram provided in an embodiment of this application, where the dots represent the pedestrian's movement trajectory. Then, reinforcement learning is used to train the robot's control strategy. The robot's motion model adopts an omnidirectional wheel model, and the rectangles interspersed among the dots represent the robot's movement trajectory. As the navigation environment changes, the robot also continuously changes its shape and size to increase the navigation model's generalization ability to different size configurations. After training, the navigation model is applied to a real robot. It should be noted that the robot's specific shape is represented by rectangles and circles, with the rectangle's length and width ranging from [0.3, 0.6], and the circle's radius ranging from [0.3, 0.6]. Furthermore, an asynchronous proximal policy optimization algorithm is used to train the navigation model, which significantly increases the policy convergence speed. Its navigation speed for the robot is expressed as linear velocity v. robot ∈[0, 0.6], angular velocity w robot∈[-0.9, 0.9], using linear velocity vm / s and angular velocity w rad / s in continuous space.

[0061] Furthermore, in reinforcement learning, laser information coding and pedestrian information coding are used to construct the coded information of the environment to better achieve simulation. See also Figure 3 This figure shows the logarithmic map and pedestrian map provided in the embodiments of this application. Laser information encoding: Laser information includes instantaneous static environmental information. First, the distance value returned by each original 2D planar laser beam is subtracted from the size of the robot's center distance contour at the current laser beam angle to incorporate the robot contour information. Then, the laser beam is downsampled to compress the number of laser beams, and concentric circles are cut along the laser length. Laser distances within the cut distance interval are set to 1.0, laser distances less than the lower limit of the cut distance are set to 0.5, and laser distances greater than the upper limit of the cut distance are set to 0. The cut interval gradually increases according to an exponential function, finally obtaining a two-dimensional vector, called the logarithmic map. Pedestrian information encoding: Pedestrian information is defined using the position and relative velocity of pedestrians and the robot. To better describe the spatial nature of pedestrian position information, pedestrian information is also mapped to a pedestrian map of the same size as the sensor map. Then, a three-channel pedestrian map is used to represent pedestrian information. One channel represents the pedestrian's position, and the other two channels represent the pedestrian's x-axis and y-axis velocities. Finally, the radius of the robot's circumcircle is used as the expansion radius to expand the radius of the pedestrian in the map, which is called the pedestrian map. Thus, the logarithmic map and the pedestrian map are merged into a four-channel map, which serves as the encoded information of the environment and can be used as input information for building the navigation model.

[0062] S102: Generate the robot's path using the navigation model according to the navigation instructions, and adjust the robot's shape and size.

[0063] In this step, the navigation model constructed above is used to generate the robot's route in the navigation environment between the client and the robot. Then, the shape and size of the robot are adjusted based on the number of people and their trajectories.

[0064] Furthermore, the technical solution of this application also utilizes a reward function to control the robot's movement based on a path. Specifically, the reward function is as follows:

[0065] r = r reach +r col +r step +r shaping

[0066]

[0067]

[0068] Where r is the sum of the rewards of the navigation model, r col As a penalty for the robot encountering obstacles during movement, r reach This indicates that the robot is encouraged to navigate from the starting point to the destination point, and can be expressed as 1000, r. step To encourage the robot to reach the target point in the shortest possible time, a small penalty of -5 can be applied. shaping For the induced arrival term, α col This indicates an encouragement for the robot to avoid obstacles during navigation, and a value of -500 can be assigned. This represents the distance between the robot and the obstacle at time t-1. Let r be the distance between the robot and the obstacle at time t, so r shaping It is represented as the distance between the robot and the previous step minus the current distance. ε1 represents encouraging the robot to continuously explore and move closer to the target point, and can be 200.

[0069] As can be seen, this optional solution mainly describes how to improve navigation efficiency. Specifically, in this optional solution, the application first obtains navigation instructions, then generates the robot's path using a navigation model based on the navigation instructions, and adjusts the robot's shape and size.

[0070] In summary, this embodiment incorporates the robot's shape and size into the navigation model. After receiving navigation instructions, the navigation model, trained through reinforcement learning, will automatically generate the robot's route and adaptively adjust the robot's shape and size to better avoid obstacles. This eliminates the need for manual parameter adjustment, thus improving both navigation efficiency and safety.

[0071] Device Examples

[0072] The navigation device for adjusting the shape and size of a robot provided in the embodiments of this application will be described below. The navigation device for adjusting the shape and size of a robot described below can be referred to in correspondence with the navigation method for adjusting the shape and size of a robot described above.

[0073] See Figure 4 This figure is a schematic diagram of the structure of the navigation device for adjusting the shape and size of a robot provided in an embodiment of this application, such as... Figure 4 As shown, the device includes:

[0074] Module 100 is used to acquire navigation instructions;

[0075] The generation module 200 is used to generate the robot's path using a navigation model according to the navigation instructions, and to adjust the shape and size of the robot.

[0076] Optional, also includes:

[0077] The first acquisition subunit is used to acquire the number of people and their trajectories corresponding to the current navigation environment;

[0078] The generation module includes:

[0079] An adjustment module is used to adjust the shape and size of the robot based on the number of people and their trajectory.

[0080] Optional, also includes:

[0081] The second acquisition subunit is used to acquire multiple different navigation environments and the number of people and their trajectories corresponding to the multiple different navigation environments.

[0082] Optionally, the construction process of the navigation model is as follows:

[0083] The third acquisition subunit is used to acquire the target navigation environment, and determine the sample navigation label corresponding to the target navigation environment based on the number of target personnel and the trajectory of the target personnel. The sample navigation label is used to determine the robot's path and shape size corresponding to the navigation command.

[0084] The acquisition module is used to input the target navigation environment into the initial navigation model to obtain the navigation label to be determined;

[0085] An adjustment module is used to adjust the initial navigation model based on the difference between the sample navigation label and the undetermined navigation label, thereby obtaining the navigation model.

[0086] Optional, also includes:

[0087] A control module is used to control the robot to move based on the path using a reward function.

[0088] The navigation device for adjusting the shape and size of a robot provided in this application has the same beneficial effects as the navigation method for adjusting the shape and size of a robot provided in the above embodiments, and therefore will not be described again.

[0089] Electronic device examples

[0090] See Figure 5 This figure is a schematic diagram of an electronic device structure provided in an embodiment of this application, such as... Figure 5 As shown, it includes:

[0091] Memory 11 is used to store computer programs;

[0092] The processor 12 is used to implement the steps of the navigation method for adjusting the shape and size of the robot as described in any of the above method embodiments when executing the computer program.

[0093] In this embodiment, the device can be an in-vehicle computer, a PC (Personal Computer), or a terminal device such as a smartphone, tablet computer, handheld computer, or portable computer.

[0094] The device may include a memory 11, a processor 12, and a bus 13. The memory 11 includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic storage, magnetic disk, optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the device, such as the hard disk of the device. In other embodiments, the memory 11 may be an external storage device of the device, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 may include both internal and external storage units of the device. The memory 11 can be used not only to store application software and various types of data installed on the device, such as program code for executing fault prediction methods, but also to temporarily store data that has been output or will be output.

[0095] In some embodiments, processor 12 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 11 or process data, such as program code for executing a fault prediction method.

[0096] This bus 13 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0097] Furthermore, the device may also include a network interface 14, which may optionally include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), typically used to establish communication connections between the device and other electronic devices.

[0098] Optionally, the device may further include a user interface 15, which may include a display, an input unit such as a keyboard, and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the device and to display a visual user interface.

[0099] Figure 5 Only devices with components 11-15 are shown; those skilled in the art will understand that... Figure 5 The structure shown does not constitute a limitation on the device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0100] Readable storage medium embodiments

[0101] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps of the navigation method for adjusting the shape and size of a robot as described in any of the above method embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0102] It should be noted that the navigation method for adjusting the shape and size of a robot provided by this invention can be used in the field of robot navigation. The above is merely an example and does not limit the application area of ​​the navigation method for adjusting the shape and size of a robot provided by this invention.

[0103] It should also be noted that the "first" and "second" in the names such as "first" and "second" (if they exist) mentioned in the embodiments of this application are only used as name identifiers and do not represent the first and second in order.

[0104] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0105] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0106] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0107] The navigation method, apparatus, device, and storage medium for adjusting the shape and size of a robot provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A navigation method for adjusting the shape and size of a robot, characterized in that, include: Get navigation instructions; Based on the navigation instructions, the robot's path is generated using a navigation model, and the robot's shape and size are adjusted. Before obtaining navigation instructions, the following is also included: Acquire multiple different navigation environments and the corresponding number of people and their trajectories under each of the multiple different navigation environments; The target navigation environment is any one of the multiple different navigation environments; the target number of personnel is any one of the corresponding personnel numbers in the multiple different navigation environments; and the target personnel trajectory is any one of the corresponding personnel trajectories in the multiple different navigation environments. The robot's path and shape / size differ in different navigation environments. The construction process of the navigation model is as follows: The target navigation environment is obtained, and sample navigation labels corresponding to the target navigation environment are determined based on the number of target personnel and the trajectory of the target personnel. The sample navigation labels are used to determine the robot's path and shape size corresponding to the navigation command. The target navigation environment is input into the initial navigation model to obtain the navigation label to be determined; Based on the difference between the sample navigation labels and the undetermined navigation labels, the initial navigation model is adjusted to obtain the navigation model.

2. The method according to claim 1, characterized in that, The navigation instructions include the current navigation environment, and after obtaining the navigation instructions, the following are also included: Obtain the number of people and their trajectories corresponding to the current navigation environment; Adjusting the shape and size of the robot includes: The shape and size of the robot are adjusted based on the number of people and their trajectories.

3. The method according to claim 1, characterized in that, Also includes: The robot is controlled to move based on the path by using a reward function.

4. A navigation device for adjusting the shape and size of a robot, characterized in that, include: The acquisition module is used to acquire navigation instructions; The generation module is used to generate the robot's path using a navigation model according to the navigation instructions, and to adjust the robot's shape and size; The second acquisition subunit is used to acquire multiple different navigation environments and the number of people and their trajectories corresponding to the multiple different navigation environments; The target navigation environment is any one of the multiple different navigation environments; the target number of personnel is any one of the corresponding personnel numbers in the multiple different navigation environments; and the target personnel trajectory is any one of the corresponding personnel trajectories in the multiple different navigation environments. The robot's path and shape / size differ in different navigation environments. The construction process of the navigation model is as follows: The target navigation environment is obtained, and sample navigation labels corresponding to the target navigation environment are determined based on the number of target personnel and the trajectory of the target personnel. The sample navigation labels are used to determine the robot's path and shape size corresponding to the navigation command. The target navigation environment is input into the initial navigation model to obtain the navigation label to be determined; Based on the difference between the sample navigation labels and the undetermined navigation labels, the initial navigation model is adjusted to obtain the navigation model.

5. The apparatus according to claim 4, characterized in that, Also includes: The first acquisition subunit is used to acquire the number of people and their trajectories in the current navigation environment. The generation module includes: An adjustment module is used to adjust the shape and size of the robot based on the number of people and their trajectory.

6. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the navigation method for adjusting the shape and size of a robot as described in any one of claims 1 to 3 when executing the computer program.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the navigation method for adjusting the shape and size of a robot as described in any one of claims 1 to 3.

Images