Unmanned aerial vehicle voice control system and method

Abstract

The invention discloses an unmanned aerial vehicle voice control system and method. The system comprises a voice acquisition module used for acquiring a voice instruction sent by an unmanned aerial vehicle operator; the voice processing module is used for performing preprocessing and voice recognition on the voice instruction received by the voice acquisition module, and converting the voice instruction into text data based on a voice recognition result; the unmanned aerial vehicle ground control module is used for generating an instruction flow based on the text data and an instruction library and sending the instruction flow to an unmanned aerial vehicle operator for confirmation and sending; the instruction library is used for storing keywords of control instructions; the unmanned aerial vehicle airborne computer is used for receiving the instruction process and controlling the unmanned aerial vehicle to act based on the instruction process; and the communication module is used for communication between the unmanned aerial vehicle ground control module and the unmanned aerial vehicle airborne computer. According to the invention, accurate control of actions of the unmanned aerial vehicle by voice is realized, the operation control process of the unmanned aerial vehicle is greatly simplified, and the control efficiency is remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles, and more specifically, relates to a voice control system and method for unmanned aerial vehicles. Background Art

[0002] With the continuous development of drone technology, its application scenarios are becoming more and more extensive, such as aerial photography, logistics distribution, agricultural plant protection, etc. The traditional drone control method is mainly manual operation through a remote control. This method may cause inconvenience and operational difficulties in some complex environments or when operators need to perform multiple tasks at the same time. Therefore, the development of a more convenient and efficient drone control method has become a current research hotspot. The emergence of voice control technology provides a new solution for drone control. The present invention aims to provide a stable, accurate and fully functional voice control system to achieve precise control of drones.

[0003] The information disclosed in the background technology section of the present invention is only intended to deepen the understanding of the general background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0004] The purpose of the present invention is to propose a UAV voice control system and method, so as to solve the problem that traditional UAV control methods may cause inconvenience and operational difficulties in some complex environments or when operators are required to perform multiple tasks at the same time, realize voice control of UAV actions, simplify the operation control of UAV, and improve control efficiency.

[0005] To achieve the above objectives, the present invention proposes a UAV voice control system and method.

[0006] According to a first aspect of the present invention, a drone voice control system is proposed, comprising:

[0007] Voice collection module, used to collect voice commands issued by drone operators;

[0008] A voice processing module, used for preprocessing and voice recognition of the voice instructions received by the voice acquisition module, and converting the voice instructions into text data based on the voice recognition results;

[0009] The UAV ground control module is used to generate an instruction flow based on the text data and the instruction library, and send the instruction flow to the UAV operator for confirmation and sending;

[0010] Instruction library, used to store keywords of control instructions;

[0011] The drone onboard computer is used to receive the instruction flow and control the drone action based on the instruction flow;

[0012] The communication module is used for communication between the UAV ground control module and the UAV onboard computer.

[0013] Optionally, the voice collection module includes:

[0014] Recording equipment and microphones on the ground side.

[0015] Optionally, the preprocessing includes:

[0016] Amplification, noise reduction, filtering and feature parameter extraction.

[0017] Optionally, the voice processing module uses a voice recognition algorithm to recognize the voice instruction.

[0018] Optionally, the speech recognition algorithm includes:

[0019] Neural network algorithms.

[0020] Optionally, the UAV ground control module matches the text data with keywords corresponding to control instructions in the instruction library to obtain corresponding control instructions, and generates a corresponding instruction flow based on the order of each of the control instructions.

[0021] Optionally, the drone operator confirms and sends the command process through voice, keyboard, mouse or touch.

[0022] Optionally, the instruction library is a multi-model instruction library, storing different types of instruction keywords and corresponding operation instruction sets for different models of drones.

[0023] Optionally, the instruction library includes:

[0024] A command mapping table for each model of drone, which records the keywords of all voice commands that each model of drone can respond to and their corresponding operation commands, and sets reasonable priority levels for different types of operation commands.

[0025] According to a second aspect of the present invention, a voice control method is proposed, based on the drone voice control system according to any one of the first aspects, comprising:

[0026] Collect voice commands from drone operators;

[0027] Preprocessing and voice recognition are performed on the voice command received by the voice acquisition module, and converting the voice command into text data based on the voice recognition result;

[0028] Generate an instruction flow based on the text data and the instruction library, and send the instruction flow to the drone operator for confirmation and sending;

[0029] Receive the instruction flow, and control the drone action based on the instruction flow.

[0030] The beneficial effects of the present invention are as follows: the present invention collects voice commands issued by a drone operator through a voice collection module, pre-processes and performs voice recognition on the voice commands received by the voice collection module through a voice processing module, and converts the voice commands into text data based on the voice recognition results; generates a command flow according to the text data and a command library through a drone ground control module, and sends the command flow to the drone operator for confirmation and to the drone onboard computer; controls the drone action according to the command flow through the drone onboard computer, realizes voice control of the drone action, simplifies the operation control of the drone, and improves the control efficiency.

[0031] The system of the present invention has other characteristics and advantages, which will be apparent from the drawings incorporated herein and the following detailed description, or will be described in detail in the drawings incorporated herein and the following detailed description, which together serve to explain the specific principles of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, in which like reference numerals generally represent like components.

[0033] Figure 1 A schematic diagram of a drone voice control system according to Embodiment 1 of the present invention is shown. DETAILED DESCRIPTION

[0034] The present invention will be described in more detail below with reference to the accompanying drawings. Although preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0035] A drone voice control system according to the present invention comprises:

[0036] Voice collection module, used to collect voice commands issued by drone operators;

[0037] The voice processing module is used to pre-process and perform voice recognition on the voice instructions received by the voice acquisition module, and convert the voice instructions into text data based on the voice recognition results;

[0038] The UAV ground control module is used to generate a command flow based on text data and a command library, and send the command flow to the UAV operator for confirmation and sending;

[0039] Instruction library, used to store keywords of control instructions;

[0040] The UAV onboard computer is used to receive the command flow and control the UAV action based on the command flow;

[0041] The communication module is used for communication between the UAV ground control module and the UAV onboard computer.

[0042] Specifically, in the operation process of the entire voice control system, the voice acquisition module plays a role first. The voice acquisition module can be equipped with a high-sensitivity microphone array. These microphones have been carefully adjusted to accurately capture the voice commands issued by the drone operator in various complex environments. Whether in a noisy outdoor work site or in a relatively closed indoor test environment, the microphone can effectively suppress environmental noise and focus on the operator's voice signal. When the operator issues a clear and unambiguous voice command, such as "the drone rises to a height of 50 meters, keeps hovering and turns on the camera to shoot", the voice acquisition module will immediately convert this sound signal into an electrical signal, and transmit it stably to the voice processing module through an optimized signal transmission line.

[0043] After receiving the voice command from the voice acquisition module, the voice processing module will use a series of advanced preprocessing technologies. First, the high-frequency noise and low-frequency interference in the signal are removed by filters to make the voice signal purer. Then, the feature extraction algorithm is used to extract the key features in the voice signal, such as Mel-frequency cepstral coefficients (MFCC), etc. These features will serve as an important basis for subsequent voice recognition. In the voice recognition link, a deep learning-based voice recognition model is used. This model has been trained with a large number of different voice samples and can accurately convert the preprocessed voice signal into the corresponding text data. For example, for the voice command mentioned above, the voice processing module can accurately recognize and convert it into the text information of "the drone rises to a height of 50 meters, keeps hovering and turns on the camera to shoot."

[0044] Subsequently, the ground control module of the UAV starts to work. It will quickly obtain the text data transmitted by the voice processing module, and compare and match it with the instruction information pre-stored in the instruction library. The instruction library records in detail the standard instruction formats and related parameter settings of various UAV operations, covering flight attitude adjustment, altitude and speed control, mission equipment operation and other aspects. The ground control module generates a complete and accurate instruction flow based on the key information in the text data, such as "ascend", "50 meters", "hover", "turn on the camera", etc., combined with the instruction flow template in the instruction library. This set of instruction flow not only includes detailed operation steps for each action of the UAV, but also takes into account the sequence and time interval between actions to ensure the safe, stable and efficient operation of the UAV. After generating the instruction flow, the ground control module will present it to the UAV operator in an intuitive and clear manner, and the operator can carefully check the accuracy and completeness of the instruction on the interface. After confirmation, the operator clicks the confirmation button, and the ground control module will send the instruction flow to the UAV onboard computer through a stable and reliable wireless communication link.

[0045] As the "brain" of the drone, the onboard computer of the drone quickly starts the command parsing and execution program after receiving the command flow sent by the ground control module. It first parses the command flow in detail and decomposes it into a series of specific control instructions, such as controlling the motor speed to achieve the rising height, adjusting the flight control parameters to maintain the hovering state, sending a signal to turn on the camera, etc. Then, according to these parsed instructions, the various execution components and systems of the drone are accurately controlled. For example, when receiving the rising command, the onboard computer sends the corresponding control signal to the motor drive module to adjust the motor speed and torque so that the drone can rise smoothly to the specified height of 50 meters; in the hovering stage, it continuously receives and analyzes the attitude data from the flight control system, and continuously fine-tunes the output of the motor through the closed-loop control algorithm to ensure that the drone maintains a stable hover at the target position; when the camera needs to be turned on, it sends a start signal to the image acquisition system, and configures the relevant shooting parameters, such as resolution, frame rate, shooting angle, etc., to meet the requirements of the shooting task. Through such a series of precise and efficient control actions, accurate voice control of the UAV's movements is achieved, which greatly simplifies the UAV's operational control process and significantly improves the control efficiency. The UAV can respond quickly and accurately to the operator's voice commands in various complex mission scenarios, thereby improving the overall operational efficiency and user experience.

[0046] In one example, the voice collection module includes:

[0047] Recording equipment and microphones on the ground side.

[0048] In one example, preprocessing includes:

[0049] Amplification, noise reduction, filtering and feature parameter extraction.

[0050] Specifically, after the voice processing module receives the voice command from the voice acquisition module, it will use a series of advanced preprocessing technologies. First, the collected voice signal is amplified and the weak electrical signal is enhanced to a suitable amplitude range through the amplifier to ensure that the subsequent processing link can more accurately extract the signal features. Then, the noise reduction algorithm is used to eliminate the interference of environmental noise on the voice signal. In this process, advanced signal processing technologies such as adaptive filtering and spectral subtraction are used to subtract the noise component from the mixed signal through real-time monitoring and analysis of environmental noise, thereby restoring a pure voice signal. Then, the signal is filtered using a filter to remove high-frequency noise and low-frequency interference in the signal, further optimizing the quality of the voice signal. After this series of amplification, noise reduction and filtering, the voice signal is purer and more stable, providing high-quality input data for the subsequent voice recognition link. Feature parameter extraction is a key link in voice processing, and its purpose is to extract key parameter information that can effectively characterize voice characteristics from the original voice signal. Through this process, complex and redundant speech waveform data can be converted into a set of concise and representative feature parameter sequences. These parameters can more prominently reflect the essential characteristics of speech, thereby providing a high-quality data foundation for subsequent speech recognition, speech synthesis and other tasks, greatly improving processing efficiency and accuracy, and reducing the complexity of data storage and calculation.

[0051] In one example, the speech processing module uses a speech recognition algorithm to recognize the speech command.

[0052] In one example, a speech recognition algorithm includes:

[0053] Neural network algorithms.

[0054] Specifically, in the speech recognition link, a speech recognition model based on a neural network algorithm is used. The neural network model has a complex structure and sophisticated design, including multiple hidden layers, such as a multi-layer perceptron (MLP) structure. Through the interconnection and weight adjustment between a large number of neurons, the input speech features are deeply abstracted and pattern recognized. In the training stage, the model uses a large amount of speech sample data containing various accents, speaking speeds, intonations, and different vocabulary combinations. The connection weights between neurons are continuously adjusted through the back propagation algorithm to learn and capture various characteristic patterns and semantic information in the speech signal. For example, for the voice command "rise" issued by different people, although there are differences in pitch, timbre, speaking speed, etc., after a long period of training, the neural network model can accurately identify its corresponding semantics and convert it into unified text information. In this way, when receiving the pre-processed voice signal, the model can quickly and accurately convert it into the corresponding text data. For example, for the above-mentioned voice command, the voice processing module can accurately recognize and convert it into the text information of "the drone rises to a height of 50 meters, keeps hovering and turns on the camera to shoot".

[0055] In one example, the UAV ground control module matches the text data with keywords corresponding to the control instructions in the instruction library to obtain the corresponding control instructions, and generates the corresponding instruction flow based on the order of the control instructions.

[0056] Specifically, after receiving the text data from the voice processing module, the drone ground control module quickly starts its core command matching and process generation mechanism. The module's built-in intelligent matching system analyzes the text data word by word and accurately matches it with the keywords corresponding to various control commands pre-stored in the command library.

[0057] The keywords in the instruction library have been carefully designed and classified, covering all dimensions of drone flight, including but not limited to flight attitude (such as "ascend", "descend", "turn left", "turn right", "lean forward", "lean back", etc.), position movement ("fly forward X meters", "translate to the left Y meters", "rise to Z height", etc.), speed control ("accelerate to XX meters per second", "decelerate to YY meters per second", etc.), mission equipment operation ("turn on the camera", "turn off the spraying system", "drop supplies", etc.) and flight mode switching ("switch to automatic return mode", "enter fixed-point hovering mode", etc.).

[0058] When the words in the text data successfully match the keywords in the instruction library, for example, "rise to 50 meters, keep hovering and turn on the camera to shoot" appears in the text, the ground control module will first identify the keywords such as "rise", "50 meters", "hover", "turn on the camera", and quickly extract the corresponding precise control instructions from the instruction library. These control instructions not only include the specific operation signals for each executive component of the drone, but also include the corresponding parameter settings, such as the target height value for the rise, the stable control parameters for hovering, and the instruction code for turning on the camera and the preset related shooting parameters.

[0059] After obtaining each matching control command, the ground control module will generate a complete, orderly and accurate command process based on the inherent logical order of these commands and the safety and rationality of drone operations. The generation process of this process fully considers the synergy and time sequence between the drone's actions, ensuring that the drone can smoothly, efficiently and safely complete various operations when executing voice commands. For example, in the above example, the command process will first plan the power output and attitude adjustment steps for the drone to rise to an altitude of 50 meters, then set the flight control parameters required to enter a stable hovering state after reaching the target altitude, and finally arrange the operation sequence of turning on the camera and configuring the shooting parameters. At the same time, the transition time and condition judgment between each step will be reasonably set to deal with various possible situations, such as automatic adjustment strategies when encountering airflow interference during the ascent process.

[0060] The generated instruction flow will be presented in an intuitive and visual way on the operating interface of the ground control module for the UAV operator to carefully check and confirm. The operator can clearly see the detailed information of each step and the overall operation process to ensure the accuracy and completeness of the instructions meet the expected flight mission requirements. Once the operator confirms that it is correct, by clicking the confirmation button on the interface, the ground control module will send this carefully generated instruction flow to the UAV onboard computer through a stable and reliable wireless communication link, thereby starting the actual operation execution phase of the UAV.

[0061] In one example, the drone operator confirms and sends the command flow via voice, keyboard, mouse, or touch.

[0062] In one example, the instruction library is a multi-model instruction library that stores different types of instruction keywords and corresponding operation instruction sets for different models of drones.

[0063] Specifically, the command library, as a key component of the entire voice control system, is highly versatile and flexible. It is designed as a multi-model command library architecture, aiming to seamlessly adapt to a variety of different models of drones to meet the needs of diverse application scenarios.

[0064] In this multi-model instruction library, an independent and detailed instruction storage system is built for each specific model of drone. It not only covers the basic instructions for drone flight control, such as common "take-off", "landing", "hovering", "forward", "backward", "turn left", "turn right" and other basic action instructions, but also stores the corresponding specific instruction keywords and corresponding operation instruction sets for the unique functions and characteristics of different models of drones.

[0065] Taking a certain type of UAV with high-precision mapping function as an example, the instruction library stores a series of command keywords specifically used for mapping operations, such as "start mapping task", "set mapping area range (X1, Y1, X2, Y2)", "adjust mapping resolution to XX PPI", "collect geographic coordinate data and save", and other complex operation instruction sets closely related to them. These operation instruction sets accurately define the specific actions and parameter settings that each system component (such as flight control system, mapping sensor, data storage module, etc.) should perform when the UAV performs mapping tasks, ensuring that the required mapping functions can be accurately realized and high-quality mapping data can be obtained.

[0066] Similarly, for a logistics drone focused on express delivery, the instruction library includes logistics distribution-specific instructions and detailed operation process instructions such as "open the cargo door", "load cargo to XX weight", "plan the transport route to the destination (latitude and longitude coordinates)", and "execute automatic delivery tasks". From the preparation for loading cargo to the route planning during transportation, and to the final delivery execution, each link has a corresponding precise instruction set, ensuring the efficient and accurate completion of logistics transportation tasks.

[0067] In order to achieve rapid identification and command matching for different types of drones, the command library adopts an intelligent identification and loading mechanism. When the system is connected to a specific model of drone, the command library will first read the model identification information sent by the drone, and then quickly locate and load the corresponding exclusive command set. This automatic adaptation process ensures that the system can quickly switch between different types of drones and accurately execute the corresponding voice control commands without manual intervention to switch and adjust the command library, greatly improving the system's ease of use and operational efficiency.

[0068] In addition, the multi-model instruction library also has good scalability and update mechanism. With the continuous development of drone technology and the launch of new models, the instruction library can easily add, modify and optimize instructions. Whether it is due to the expansion of new functional instructions brought about by drone hardware upgrades, or to meet the personalized instruction needs of specific users in complex application scenarios, system administrators can easily update and maintain the instruction library through a dedicated instruction library management interface to ensure that the instruction library always keeps pace with the latest drone technology and user needs, providing continuous, stable and comprehensive instruction support for the voice control system.

[0069] Through such a powerful and well-designed multi-model command library, the voice control system can break down the barriers between drone models and achieve accurate and efficient voice control of a variety of different types of drones, providing a solid technical foundation and strong operational guarantee for the widespread application of drones in various fields.

[0070] In one example, the instruction library includes:

[0071] The command mapping table for each model of drone records the keywords of all voice commands that each model of drone can respond to and their corresponding operation commands, and sets reasonable priority levels for different types of operation commands.

[0072] Specifically, the command mapping table, as the core data structure of the multi-model command library, has been carefully constructed and optimized for each model of drone. It records the keywords of all voice commands that a specific model of drone can respond to and their corresponding precise operation instructions in a highly structured and easy-to-query and manage manner, and assigns reasonable priority levels to different types of operation instructions to ensure that the drone can operate in an orderly manner when performing complex tasks, and to maximize the flight safety and smooth completion of the task. Among these operation instructions, instructions related to drone flight safety are given the highest priority, such as the "landing" instruction. Regardless of what other tasks or operations the drone is currently performing, once the "landing" instruction is received, the flight control system will immediately interrupt all current operations and give priority to the landing procedure to ensure that the drone can land safely and avoid safety accidents caused by delays in other operations. The second is instructions related to flight attitude control and position maintenance, such as "hover", "rotate left X degrees", "rotate right X degrees", "rise X meters", "descend X meters", "fly forward X meters", "fly backward X meters", "translate to the left X meters", "translate to the right X meters", etc. These instructions have a higher priority because they directly affect the flight stability and safety of the drone. When performing other non-critical tasks (such as adjusting shooting parameters, etc.), if these attitude control and position adjustment instructions are received, the flight control system will suspend the current non-critical tasks and give priority to executing these instructions to ensure that the flight attitude and position of the drone can be adjusted in a timely and accurate manner to avoid danger caused by attitude loss or position offset. The third is the instructions related to the task function, such as "switch to automatic follow mode", "start camera shooting", "switch camera angle to overhead shooting", "switch camera angle to upward shooting", etc. The priority of these instructions is relatively low because they are mainly for realizing the specific task functions of the drone, rather than directly affecting flight safety. When the drone is in a stable flight state and no higher priority instructions are received, the flight control system will execute these task function instructions in sequence to meet the user's shooting or other task requirements. Through such a detailed instruction mapping table design, each model of drone can accurately and efficiently respond to voice instructions, realize a variety of operating functions, and reasonably arrange the operation sequence according to the priority level of the instructions in a complex task environment to ensure flight safety and the smooth completion of the task, providing users with a more convenient and intelligent drone voice control experience.

[0073] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments, but is not intended to be limiting of the present invention. It should be noted that the embodiments and features in the embodiments of the present invention may be combined with each other without conflict.

[0074] Embodiment 1

[0075] like Figure 1As shown, this embodiment provides a drone voice control system, including:

[0076] Voice collection module, used to collect voice commands issued by drone operators;

[0077] The voice processing module is used to pre-process and perform voice recognition on the voice instructions received by the voice acquisition module, and convert the voice instructions into text data based on the voice recognition results;

[0078] UAV ground control software, used to generate command flow based on text data and multi-model command library, and send the command flow to the UAV operator for confirmation and sending;

[0079] Multi-model instruction library, used to store keywords for control instructions;

[0080] The UAV onboard computer is used to receive the command flow and control the UAV action based on the command flow;

[0081] The airborne communication system and the ground communication system are used for communication between the UAV ground control software and the UAV onboard computer.

[0082] The UAV flight control and navigation system is used to provide UAV attitude data and respond to the command flow sent by the ground control software through the airborne communication system and the ground communication system.

[0083] The voice acquisition module includes:

[0084] Recording equipment and microphones on the ground side.

[0085] Preprocessing includes:

[0086] Amplification, noise reduction, filtering and feature parameter extraction.

[0087] The voice processing module uses a voice recognition algorithm to recognize voice commands.

[0088] Speech recognition algorithms include:

[0089] Neural network algorithms.

[0090] The UAV ground control software matches the text data with the keywords corresponding to the control instructions in the instruction library, obtains the corresponding control instructions, and generates the corresponding instruction flow based on the sequence of each control instruction.

[0091] The drone operator confirms and sends the command process through voice, keyboard, mouse or touch.

[0092] The multi-model instruction library stores different types of instruction keywords and corresponding operation instruction sets for different models of drones.

[0093] The multi-model instruction library includes:

[0094] The command mapping table for each model of drone records the keywords of all voice commands that each model of drone can respond to and their corresponding operation commands, and sets reasonable priority levels for different types of operation commands.

[0095] Embodiment 2

[0096] This embodiment provides a voice control method, based on the drone voice control system described in Embodiment 1, including:

[0097] Collect voice commands from drone operators;

[0098] Preprocessing and voice recognition are performed on the voice command received by the voice acquisition module, and converting the voice command into text data based on the voice recognition result;

[0099] Generate an instruction flow based on the text data and the instruction library, and send the instruction flow to the drone operator for confirmation and sending;

[0100] Receive the instruction flow, and control the drone action based on the instruction flow.

[0101] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A drone voice control system, characterized in that: include: Voice collection module, used to collect voice commands issued by drone operators; A voice processing module, used for preprocessing and voice recognition of the voice instructions received by the voice acquisition module, and converting the voice instructions into text data based on the voice recognition results; The UAV ground control module is used to generate an instruction flow based on the text data and the instruction library, and send the instruction flow to the UAV operator for confirmation and sending; Instruction library, used to store keywords of control instructions; The UAV onboard computer is used to receive the instruction flow and control the UAV action based on the instruction flow; The communication module is used for communication between the UAV ground control module and the UAV onboard computer.

2. The drone voice control system according to claim 1, characterized in that: The voice collection module comprises: Recording equipment and microphones on the ground side.

3. The drone voice control system according to claim 1, characterized in that: The pre-processing comprises: Amplification, noise reduction, filtering and feature parameter extraction.

4. The drone voice control system according to claim 1, characterized in that: The voice processing module uses a voice recognition algorithm to recognize the voice instruction.

5. The drone voice control system according to claim 4, characterized in that: The speech recognition algorithm includes: Neural network algorithms.

6. The voice control system of the drone according to claim 1, characterized in that: The UAV ground control module matches the text data with keywords corresponding to the control instructions in the instruction library to obtain corresponding control instructions, and generates corresponding instruction flows based on the sequence of the control instructions.

7. The drone voice control system according to claim 1, characterized in that: The drone operator confirms and sends the command process through voice, keyboard, mouse or touch.

8. The drone voice control system according to claim 1, characterized in that: The instruction library is a multi-model instruction library that stores different types of instruction keywords and corresponding operation instruction sets for different models of drones.

9. The drone voice control system according to claim 8, characterized in that: The instruction library includes: A command mapping table for each model of drone, which records the keywords of all voice commands that each model of drone can respond to and their corresponding operation commands, and sets reasonable priority levels for different types of operation commands.

10. A voice control method, based on the drone voice control system according to any one of claims 1 to 9, characterized in that: include: Collect voice commands from drone operators; Preprocessing and voice recognition are performed on the voice command received by the voice acquisition module, and converting the voice command into text data based on the voice recognition result; Generate an instruction flow based on the text data and the instruction library, and send the instruction flow to the drone operator for confirmation and sending; Receive the instruction flow, and control the drone action based on the instruction flow.

Images