CN115562299A - Navigation method, device, mobile robot and medium of a mobile robot - Google Patents

Abstract

The invention discloses a navigation method and a navigation device for a mobile robot, the mobile robot and a medium. The method comprises the following steps: after a mobile robot executes current moving operation, inputting current sensor data of the mobile robot into a pre-trained target network model to obtain output first controller parameters; determining a second controller parameter based on the current actual track parameter and the current preset track parameter of the mobile robot; controlling the mobile robot to perform a next movement operation based on the first controller parameter and the second controller parameter. The embodiment of the invention solves the problem that the existing navigation method of the mobile robot depends on a single deviation rectifying algorithm, and improves the navigation accuracy of the mobile robot.

Description

Navigation method, device, mobile robot and medium of a mobile robot

technical field

The invention relates to the technical field of robots, in particular to a navigation method and device for a mobile robot, a mobile robot and a medium.

Background technique

With the rapid development of science and technology, a robot system with mobile functions composed of sensors and automatic controllers has become one of the hot research directions. Mobile robots are widely used in production, construction and monitoring industries.

In actual scene applications, mobile robots usually need to navigate and move according to preset trajectories, but due to the influence of the performance parameters of the mobile robot's fuselage and the external environment, there is usually a trajectory error between the actual trajectory of the mobile robot and the preset trajectory , resulting in the mobile robot being unable to accurately reach the target location to perform the task operation.

The existing navigation methods of mobile robots mainly rely on a single deviation correction algorithm, which makes the navigation accuracy of mobile robots not high.

Contents of the invention

Embodiments of the present invention provide a mobile robot navigation method, device, mobile robot and medium to solve the problem that the existing mobile robot navigation method relies on a single deviation correction algorithm and improve the navigation accuracy of the mobile robot.

According to one embodiment of the present invention, a navigation method for a mobile robot is provided, the method comprising:

After the mobile robot performs the current movement operation, input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters;

determining second controller parameters based on current actual trajectory parameters and current preset trajectory parameters of the mobile robot;

Based on the first controller parameter and the second controller parameter, the mobile robot is controlled to perform a next moving operation.

According to another embodiment of the present invention there is provided a navigation device for a mobile robot, the device comprising:

The first controller parameter determination module is used to input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters after the mobile robot performs the current movement operation;

The second controller parameter determination module is used to determine the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot;

A mobile robot control module, configured to control the mobile robot to perform a next moving operation based on the first controller parameter and the second controller parameter.

According to another embodiment of the present invention, a mobile robot is provided, and the mobile robot includes: a drive controller and at least one sensor;

Wherein, the sensor is used to collect sensor data;

The drive controller includes at least one processor; and a memory connected in communication with the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the The at least one processor is executed, so that the at least one processor can execute the navigation method of the mobile robot described in any embodiment of the present invention.

According to another embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement any of the embodiments of the present invention when executed. Navigation methods for mobile robots.

In the technical solution of the embodiment of the present invention, after the mobile robot performs the current movement operation, the current sensor data of the mobile robot is input into the pre-trained target network model to obtain the output first controller parameters, based on the current The actual trajectory parameters and the current preset trajectory parameters are used to determine the second controller parameters. Based on the first controller parameters and the second controller parameters, the mobile robot is controlled to perform the next movement operation, and the controller parameters are respectively obtained based on the two dimensions. The problem that the existing mobile robot navigation method relies on a single deviation correction algorithm is solved, and the navigation accuracy of the mobile robot is improved.

It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will be easily understood from the following description.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort.

Fig. 1 is the flowchart of the navigation method of a kind of mobile robot provided by one embodiment of the present invention;

FIG. 2 is a flowchart of another mobile robot navigation method provided by an embodiment of the present invention;

FIG. 3 is an architecture diagram of a target network model provided by an embodiment of the present invention;

4 is a flowchart of a specific example of a navigation method for a mobile robot provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a navigation device for a mobile robot provided by an embodiment of the present invention;

Fig. 6 is a schematic structural diagram of a mobile robot provided by an embodiment of the present invention;

Fig. 7 is a schematic structural diagram of a drive controller provided by an embodiment of the present invention.

detailed description

In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only It is an embodiment of a part of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms "first" and "second" in the description and claims of the present invention and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

Fig. 1 is a flow chart of a navigation method for a mobile robot provided by an embodiment of the present invention. This embodiment is applicable to the situation where a mobile robot is navigated and tracked, and the method can be executed by a navigation device of a mobile robot. The navigation device of the mobile robot can be implemented in the form of hardware and/or software, and the navigation device of the mobile robot can be configured in the mobile robot. As shown in Figure 1, the method includes:

S110. After the mobile robot performs the current movement operation, input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters.

The mobile robot in this embodiment can be a soft robot or a mechanical robot, and is especially suitable for a soft robot. Among them, soft robots are made of soft smart materials. Unlike mechanical robots, the driving method of soft robots mainly depends on the smart materials used. For example, smart materials are generally dielectric elastomers (DE), ion polymer metal composites, etc. materials (IPMC), shape memory alloys (SMA), shape memory polymers (SMP) and more. The smart materials used in the soft robot are not limited here.

Since mechanical robots can rely on traditional mechanical algorithms to construct mechanical or kinematic models, most of the existing technologies calculate the controller parameters of mechanical robots directly based on sensor data, but the driving characteristics of soft robots are not suitable for traditional mechanical algorithms, sensor data and control. There is no curve law that can be fitted to the sensor parameters. Accordingly, the embodiment of the present invention proposes a method of using the target network model to simulate the correlation law between the sensor data and the controller parameters.

Wherein, for example, the current sensor data includes but not limited to the current speed of the three axes, the current acceleration of the three axes, the current angular velocity of the three axes, the current angle of the three axes, the current temperature, the current wind speed, etc., wherein the three axes include x-axis, y-axis and z-axis. The current sensor data collected by the mobile robot is not limited here, and the user can set the required sensors on the mobile robot according to actual needs.

Wherein, exemplary, the model architecture of target network model includes but not limited to CNN network (ConvolutionalNeural Networks, convolutional neural network), FCN network (Fully Convolutional Networks, fully convolutional neural network), residual network (ResNet), DNN network (Deep Neural Networks, deep neural network), RNN network (Recurrent Neural Network, cyclic neural network), multi-channel LSTM network (Long Short-Term Memory, long-short-term memory) or Transformer network, etc., here the target network model The model architecture is not limited.

Among them, exemplary, the first controller parameters include but not limited to the amplitude of the current, the frequency of the current, the duty cycle of the current, the duration of the current, the amplitude of the voltage, the frequency of the voltage, the duty cycle of the voltage and The duration of the voltage, etc., there is no limitation on the parameters of the first controller, and the user can customize the settings according to actual needs.

S120. Determine second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot.

Specifically, the current actual trajectory parameters are used to represent the actual trajectory parameters of the mobile robot after performing the current movement operation. Exemplarily, the actual trajectory parameters include but are not limited to actual position coordinates, actual heading angles, and the like. The actual trajectory parameters are not limited here.

Specifically, the current preset trajectory parameters are used to characterize the preset trajectory parameters of the mobile robot after performing the current movement operation. Exemplarily, the preset trajectory parameters include but are not limited to preset position coordinates, preset heading angles, etc. . There is no limitation to the preset track parameters here.

In an optional embodiment, determining the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot includes: using a preset closed-loop control algorithm, based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot. Set the trajectory parameters and determine the second controller parameters.

Wherein, for example, the preset closed-loop control algorithm includes but not limited to positional control algorithm, PID control algorithm, robust control algorithm, Dalin algorithm and LQR (Linear Quadratic Regulator linear quadratic regulator) algorithm and so on. The preset closed-loop control algorithm is not limited here.

On the basis of the foregoing embodiments, optionally, the method further includes: acquiring current actual trajectory parameters collected by the vision device, and/or, based on current sensor data, determining the current actual trajectory parameters.

In an optional embodiment, a vision device is installed on the mobile robot or in the scene environment to determine the current actual trajectory parameters based on the collected trajectory images.

In another optional embodiment, specifically, the current actual trajectory parameter is determined based on the last actual trajectory parameter and current sensor data.

In another optional embodiment, the average value of the current actual trajectory parameters collected by the vision device and the current actual trajectory parameters determined based on the current sensor data is used as the current actual trajectory parameters of the mobile robot.

S130. Based on the first controller parameter and the second controller parameter, control the mobile robot to perform the next moving operation.

In an optional embodiment, the average value of the first controller parameter and the second controller parameter is used as the target controller parameter, and the mobile robot is controlled to perform the next movement operation based on the target controller parameter. Wherein, specifically, the target controller parameter is used to characterize the parameter data input into the driving controller of the mobile robot. Exemplarily, the driving controller includes but not limited to a controlled power supply and a PWM module.

Wherein, for example, the voltage amplitude in the first controller parameter is 5V, the voltage amplitude in the second controller parameter is 7V, and the voltage amplitude in the target controller parameter is 6V.

In the technical solution of this embodiment, after the mobile robot performs the current movement operation, the current sensor data of the mobile robot is input into the pre-trained target network model to obtain the output first controller parameters, based on the current actual situation of the mobile robot The trajectory parameters and the current preset trajectory parameters determine the second controller parameters, control the mobile robot to perform the next movement operation based on the first controller parameters and the second controller parameters, and obtain the controller parameters based on the two dimensions respectively, which solves the problem of Existing navigation methods for mobile robots rely on the problem of a single deviation correction algorithm, which improves the navigation accuracy of mobile robots.

FIG. 2 is a flow chart of another mobile robot navigation method provided by an embodiment of the present invention. This embodiment further refines the target network model in the foregoing embodiments. As shown in Figure 2, the method includes:

S210. Input each training sensor data in the training sensor data set into the initial network model, and obtain at least one output predictive controller parameter.

Wherein, for example, the training sensor data set may be sensor data collected by the sensor during the movement of the mobile robot.

S220. Determine a loss function based on each predictive controller parameter and a standard controller parameter respectively corresponding to each predictive controller parameter.

Wherein, for example, the standard controller parameters may be parameter data input into the drive controller during the moving process of the mobile robot.

In an optional embodiment, the standard controller parameters corresponding to the current training sensor data collected at the current moment are the standard controller parameters collected at the next moment.

Wherein, for example, the loss function includes but not limited to square loss function, logarithmic loss function, exponential loss function, logistic regression loss function, Huber loss function, cross entropy loss function and Kullback-Leibler divergence loss function and so on. The loss function is not limited here.

S230. Based on the loss function, adjust the model parameters of the initial network model until the loss function converges to obtain a trained target network model.

On the basis of the above embodiments, optionally, the method further includes: inputting each verification sensor in the verification sensor data set into the target network model, obtaining at least one output predictive controller parameter, based on each predictive controller parameter and The standard controller parameters corresponding to each predictive controller parameter are used to determine the performance parameters of the target network model; when the performance parameters do not meet the preset parameter standards, the hyperparameters of the target network model are adjusted, and the adjusted The target network model is trained.

Wherein, for example, the hyperparameters include but not limited to the number of network layers, the number of network nodes, the number of iterations, the learning rate and so on.

The advantage of this setting is that it can improve the generalization ability of the trained target network model.

In this embodiment, the target network model is a multi-channel long-short-term memory model. In an optional embodiment, the target network model includes an output module and at least one long-short-term memory module, wherein the first long-short-term memory module is used to output the first long-short-term feature vector based on the input current sensor data, and the second The i long-short-term memory module is used to output the i-th long-short-term feature vector based on the input current sensor data and the i-1th long-short-term feature vector output by the i-1th long-short-term memory module, and the output module is used based on The nth long-short-term feature vector output by the nth long-short-term memory module outputs the first controller parameter; wherein, i is an integer greater than 1 and less than n, and n represents the number of long-short-term memory modules in the target network model.

In an optional embodiment, the i-th long-short-term memory module includes an input unit and a long-short-term memory unit, the input unit includes a first linear layer, a self-attention layer, and a second linear layer, and the first linear layer is used to input The current sensor data is linearly encoded to output the first sensor feature, the self-attention layer is used to output the fusion sensor data based on the first sensor feature output by the first linear layer, and the second linear layer is used to output the self-attention layer. Fusion sensor data for linear encoding, output the second sensor feature, the long-short-term memory unit is used to output the second sensor feature based on the second linear layer and the i-1th long-short-term feature output by the i-1th long-short-term memory module Vector, output the i-th long-short-time feature vector.

Among them, the self-attention layer is used to realize multi-channel data fusion. Exemplarily, the self-attention layer satisfies the formula:

Among them, Q represents the query vector matrix, K represents the key vector matrix, V represents the value vector matrix, and _dk represents the vector used for querying.

Wherein, specifically, the long-short-term memory unit includes a memory unit c, an input gate i, a forget gate f and an output gate o. Among them, the update process of the long-short-term memory unit at time t satisfies the following formula:

f _t = σ(W _f ·[h _t-1 ,x _t ]+b _f ); it =σ(W _i ·[h _{t -1} ,x _t ]+b _i );

o _t ＝σ(W _o ·[h _t-1 ,x _t ]+b _o ); h _t ＝o _t ⊙tanh(c _t )

表示记忆单元的候选记忆状态值，h_t表示长短时记忆单元输出的长短时特征向量，σ表示logistic sigmoid函数，W_f、W_i、W_c、W_o、b_f、b_i、b_c和b_o表示权重矩阵。Among them, x _t represents the input data of the long-short-term memory unit at time t, f _t , it , c _t and o _t represent the values of the forget gate, input gate, memory unit and output gate at time _t , respectively,

Indicates the candidate memory state value of the memory unit, h _t indicates the long-short-term feature vector output by the long-short-term memory unit, σ indicates the logistic sigmoid function, W _f , W _i , W _c , W _o , b _f , _bi , b _c and b _o represents the weight matrix.

Fig. 3 is a structure diagram of a target network model provided by an embodiment of the present invention. Specifically, the target network model includes an output module and long-short-term memory modules corresponding to n moments, wherein the output module includes a linear layer and an activation layer, and each long-short-term memory module includes an input unit and a long-short-term memory unit, Wherein, the input unit includes a first linear layer, a self-attention layer and a second linear layer. The first controller parameters output by the target network model include voltage parameters for voltage modulation, current parameters for current modulation, and duty cycles for PWM modulation.

S240. After the mobile robot performs the current moving operation, input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters.

S250. Determine second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot.

S260. Based on the first controller parameter and the second controller parameter, control the mobile robot to perform the next moving operation.

In an optional embodiment, controlling the mobile robot to perform the next movement operation based on the first controller parameter and the second controller parameter includes: determining the controller parameter error based on the first controller parameter and the second controller parameter ; Based on the first controller parameter and the error of the controller parameter, determine the target controller parameter, and based on the target controller parameter, control the mobile robot to perform the next movement operation.

Wherein, specifically, the controller parameter error is used to represent the proportional difference data between the first controller parameter and the second controller parameter.

Wherein, exemplary, the target controller parameter α ^* satisfies the formula:

α ^* ＝α ₁ ±λ|α ₁ -α ₂ |

Among them, α1 represents the _first controller parameter, λ represents the proportional coefficient, and _α2 represents the second controller parameter.

Specifically, when the first controller parameter is smaller than the second controller parameter, the target controller parameter is equal to the sum of the first controller parameter and the controller parameter error, and when the first controller parameter is greater than the second controller parameter In the case of parameters, the target controller parameter is equal to the difference between the first controller parameter and the controller parameter error.

In another optional embodiment, the target controller parameter is determined based on the second controller parameter and the error of the controller parameter, and based on the target controller parameter, the mobile robot is controlled to perform the next moving operation.

The advantage of this setting is that the accuracy of the target controller parameters can be further improved, thereby further improving the navigation accuracy of the mobile robot.

Fig. 4 is a flow chart of a specific example of a navigation method for a mobile robot provided by an embodiment of the present invention, which is illustrated by taking the mobile robot as a soft robot as an example. Specifically, the soft robot is provided with a drive controller, sensors and vision devices.

Among them, after the soft robot performs the current moving operation and before reaching the end point, the current actual trajectory data collected by the visual device is sent to the PID algorithm module, so that the PID algorithm module can determine based on the current preset trajectory parameters and the current actual trajectory parameters. Second controller parameter α ₂ . The current sensor data collected by the sensor is input into the target network model to obtain the output first controller parameter α ₁ . The target controller parameter α ^* determined based on the first controller parameter α ₁ and the second controller parameter α ₂ is input into the drive controller on the soft robot, so that the drive controller controls the soft robot to perform the next movement operation.

In the technical solution of this embodiment, by inputting each training sensor data in the training sensor data set into the initial network model, at least one output predictive controller parameter is obtained, based on each predictive controller parameter and corresponding to each predictive controller parameter The standard controller parameters, determine the loss function, adjust the model parameters of the initial network model based on the loss function, until the loss function converges, and get the target network model after training, which solves the training problem of the target network model. The channel LSTM model is used as the target network model, which ensures the adaptability of the target network model and the navigation scene of the mobile robot, and further improves the navigation accuracy of the mobile robot.

Fig. 5 is a schematic structural diagram of a navigation device for a mobile robot provided by an embodiment of the present invention. As shown in FIG. 5 , the device includes: a first controller parameter determination module 310 , a second controller parameter determination module 320 and a mobile robot control module 330 .

Wherein, the first controller parameter determination module 310 is used to input the current sensor data of the mobile robot into the pre-trained target network model after the mobile robot performs the current movement operation, and obtain the first output controller parameters;

The second controller parameter determination module 320 is used to determine the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot;

The mobile robot control module 330 is configured to control the mobile robot to perform the next moving operation based on the first controller parameter and the second controller parameter.

In the technical solution of this embodiment, after the mobile robot performs the current movement operation, the current sensor data of the mobile robot is input into the pre-trained target network model to obtain the output first controller parameters, based on the current actual situation of the mobile robot The trajectory parameters and the current preset trajectory parameters determine the second controller parameters, control the mobile robot to perform the next movement operation based on the first controller parameters and the second controller parameters, and obtain the controller parameters based on the two dimensions respectively, which solves the problem of Existing navigation methods for mobile robots rely on the problem of a single deviation correction algorithm, which improves the navigation accuracy of mobile robots.

On the basis of the foregoing embodiments, optionally, the device further includes:

The target network model training module is used to input each training sensor data in the training sensor data set into the initial network model to obtain at least one predictive controller parameter output;

Determining a loss function based on each predictive controller parameter and standard controller parameters respectively corresponding to each predictive controller parameter;

Based on the loss function, the model parameters of the initial network model are adjusted until the loss function converges, and the trained target network model is obtained.

On the basis of the above embodiments, optionally, the target network model includes an output module and at least one long-short-term memory module, wherein the first long-short-term memory module is used to output the first long-short-term memory module based on the input current sensor data. The feature vector, the i-th long-short-term memory module is used to output the i-1th long-short-term feature vector based on the input current sensor data and the output of the i-1-th long-short-term memory module, output the i-th long-short-term feature vector, and output The module is used to output the first controller parameter based on the nth long-short-term feature vector output by the nth long-short-term memory module; wherein, i is an integer greater than 1 and less than n, and n represents the long-short-term memory in the target network model number of modules.

On the basis of the foregoing embodiments, optionally, the i-th long-short-term memory module includes an input unit and a long-short-term memory unit, the input unit includes a first linear layer, a self-attention layer, and a second linear layer, and the first linear layer It is used to linearly encode the input current sensor data, output the first sensor feature, the self-attention layer is used to output the first sensor feature based on the first linear layer output, and output the fusion sensor data, and the second linear layer is used for self-attention The fusion sensor data output by the force layer is linearly encoded, and the second sensor feature is output, and the long-short-term memory unit is used based on the second sensor feature output by the second linear layer and the i-1th output of the i-1th long-short-term memory module. long-short-time feature vector, and output the i-th long-short-time feature vector.

On the basis of the above embodiments, optionally, the mobile robot control module 330 is specifically used for:

determining a controller parameter error based on the first controller parameter and the second controller parameter;

Based on the first controller parameter and the error of the controller parameter, a target controller parameter is determined, and based on the target controller parameter, the mobile robot is controlled to perform a next moving operation.

On the basis of the above embodiments, optionally, the second controller parameter determination module 320 is specifically used for:

A preset closed-loop control algorithm is used to determine the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot.

On the basis of the foregoing embodiments, optionally, the device further includes:

The current actual trajectory parameter determination module is configured to obtain the current actual trajectory parameters collected by the vision device, and/or, based on the current sensor data, determine the current actual trajectory parameters.

The mobile robot navigation device provided by the embodiments of the present invention can execute the mobile robot navigation method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

Fig. 6 is a schematic structural diagram of a mobile robot provided by an embodiment of the present invention. The mobile robot includes a drive controller 410 and at least one sensor 420 for collecting sensor data. FIG. 6 takes two sensors 420 installed on the mobile robot as an example for illustration. Wherein, exemplary, at least one sensor 420 includes but not limited to an odometer, a gyroscope, an inertial sensor, a pressure sensor, a temperature sensor and the like. The components shown in the embodiments of the present invention, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the inventions described and/or claimed herein.

On the basis of the above embodiments, optionally, the mobile robot further includes a vision device 430 for collecting current actual trajectory parameters. Exemplarily, the vision device 430 may be a vision sensor.

Fig. 7 is a schematic structural diagram of a drive controller provided by an embodiment of the present invention. As shown in FIG. 7 , the drive controller 410 includes at least one processor 11, and a memory connected in communication with at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory There is stored a computer program executable by at least one processor 11, which can be executed according to a computer program stored in a read only memory (ROM) 12 or loaded from a storage unit 18 into a random access memory (RAM) 13 , to perform various appropriate actions and processing. In the RAM 13, various programs and data necessary for the operation of the drive controller 410 are also stored. The processor 11 , ROM 12 , and RAM 13 are connected to each other through a bus 14 . An input/output (I/O) interface 15 is also connected to the bus 14 .

Multiple components in the drive controller 410 are connected to the I/O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, CD, etc.; and communication unit 19, such as network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the drive controller 410 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks.

Processor 11 may be various general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes various methods and processes described above, such as a navigation method of a mobile robot.

In some embodiments, the navigation method of the mobile robot may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18 . In some embodiments, part or all of the computer program may be loaded and/or installed on the drive controller 410 via the ROM 12 and/or the communication unit 19 . When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the navigation method of the mobile robot described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured in any other suitable way (for example, by means of firmware) to execute the navigation method of the mobile robot.

Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips Implemented in a system of systems (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

Embodiment 5 of the present invention also provides a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and the computer instructions are used to make the processor execute a navigation method for a mobile robot, the method comprising:

After the mobile robot performs the current mobile operation, input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters;

Determining second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot;

Based on the first controller parameter and the second controller parameter, the mobile robot is controlled to perform the next moving operation.

In the context of the present invention, a computer readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus or device. A computer readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer readable storage medium may be a machine readable signal medium. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer discs, hard drives, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

In order to provide interaction with the user, the systems and techniques described herein can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display)) for displaying information to the user. monitor); and a keyboard and pointing device (eg, a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

A computing system can include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or a cloud host. It is a host product in the cloud computing service system to solve the problems of difficult management and weak business expansion in traditional physical hosts and VPS services. defect.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, each step described in the present invention may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution of the present invention can be achieved, there is no limitation herein.

The above specific implementation methods do not constitute a limitation to the protection scope of the present invention. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A navigation method for a mobile robot, comprising: After the mobile robot performs the current movement operation, input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters; determining second controller parameters based on current actual trajectory parameters and current preset trajectory parameters of the mobile robot; Based on the first controller parameter and the second controller parameter, the mobile robot is controlled to perform a next moving operation. 2. method according to claim 1, is characterized in that, the training method of described target network model, comprises: Input each training sensor data in the training sensor data set into the initial network model to obtain at least one predictive controller parameter output; determining a loss function based on each of the predictive controller parameters and standard controller parameters respectively corresponding to each of the predictive controller parameters; Based on the loss function, the model parameters of the initial network model are adjusted until the loss function converges to obtain a trained target network model. 3. The method according to claim 1 or 2, wherein the target network model comprises an output module and at least one long-short-term memory module, wherein the first long-short-term memory module is used for current sensor data based on input , output the first long-short-term feature vector, the i-th long-short-term memory module is used to output the i-1th long-short-term feature vector based on the input current sensor data and the i-1th long-short-term memory module output, and output the i-th A long-short-term feature vector, the output module is used to output the first controller parameter based on the n-th long-short-term feature vector output by the n-th long-short-term memory module; wherein, i is an integer greater than 1 and less than n, n Indicates the number of long short-term memory modules in the target network model. 4. The method according to claim 3, wherein the ith long-short-term memory module includes an input unit and a long-short-term memory unit, and the input unit includes a first linear layer, a self-attention layer and a second A linear layer, the first linear layer is used to linearly encode the input current sensor data, and output the first sensor feature, and the self-attention layer is used to output the first sensor feature based on the first linear layer, and output Fusion sensor data, the second linear layer is used to linearly encode the fusion sensor data output by the self-attention layer, and output second sensor features, and the long short-term memory unit is used to output based on the second linear layer The second sensor feature and the i-1th long-short-term feature vector output by the i-1th long-short-term memory module, output the i-th long-short-term feature vector. 5. The method according to claim 1, wherein said controlling said mobile robot to perform a next movement operation based on said first controller parameter and said second controller parameter comprises: determining a controller parameter error based on the first controller parameter and the second controller parameter; A target controller parameter is determined based on the first controller parameter and the controller parameter error, and based on the target controller parameter, the mobile robot is controlled to perform a next moving operation. 6. The method according to claim 1, wherein said determining the second controller parameter based on the current actual trajectory parameter and the current preset trajectory parameter of said mobile robot comprises: A preset closed-loop control algorithm is used to determine the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot. 7. The method according to claim 1, further comprising: Acquire current actual trajectory parameters collected by the vision device, and/or, based on the current sensor data, determine current actual trajectory parameters. 8. A navigation device for a mobile robot, comprising: The first controller parameter determination module is used to input the current sensor data of the mobile robot into the pre-trained target network model to obtain the output first controller parameters after the mobile robot performs the current movement operation; The second controller parameter determination module is used to determine the second controller parameters based on the current actual trajectory parameters and the current preset trajectory parameters of the mobile robot; A mobile robot control module, configured to control the mobile robot to perform a next moving operation based on the first controller parameter and the second controller parameter. 9. A mobile robot, characterized in that, the mobile robot comprises: a drive controller and at least one sensor; Wherein, the sensor is used to collect sensor data; The drive controller includes at least one processor; and a memory connected in communication with the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the The at least one processor is executed, so that the at least one processor can execute the navigation method of the mobile robot described in any one of claims 1-7. 10. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method described in any one of claims 1-7 when executed. Navigation methods for mobile robots.

Images