A mechanical exoskeleton based on human-computer voice interaction and a safety control system
Through the human-computer voice interaction system, the mechanical exoskeleton system achieves convenient control without stopping the movement while in motion, solving the problem of inconvenient blind operation in existing technologies and improving operational safety and interactive experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIFULUO (GUANGDONG) TECHNOLOGY CO LTD
- Filing Date
- 2026-05-08
- Publication Date
- 2026-07-14
AI Technical Summary
Existing mechanical exoskeletons are difficult to operate blindly in scenarios where movement or hands are occupied, lack real-time dynamic data feedback, and are inconvenient and unsafe to operate.
The mechanical exoskeleton system, based on human-computer voice interaction, receives natural language commands via a wireless Bluetooth module. Combined with an intelligent control center and sensor data acquisition, it achieves voice recognition and feedback, eliminating reliance on physical buttons and providing real-time status updates.
It enables efficient and safe blind operation control without stopping in complex work scenarios, improves voice recognition accuracy and user interaction immersion, and provides real-time data feedback that is independent of visual dependence.
Smart Images
Figure CN122378652A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mechanical exoskeleton control system technology, specifically a mechanical exoskeleton and safety control system based on human-computer voice interaction. Background Technology
[0002] A mechanical exoskeleton is a wearable augmentation device designed to provide humans with additional strength and support through mechanical structures and electrical systems. It is typically made of lightweight materials and designed to fit snugly into the user's body to help them perform a variety of physical activities, especially in work environments that require lifting heavy objects, standing for long periods of time, or walking.
[0003] Mechanical exoskeletons are widely used in many fields, such as manufacturing, construction, medical rehabilitation and military. In the rehabilitation field, mechanical exoskeletons can help paralyzed or immobile patients regain their ability to walk and improve their quality of life. In the industrial field, they can improve workers' productivity and reduce occupational injuries caused by heavy physical labor.
[0004] Currently, the mode switching, power adjustment, and status acquisition of exoskeletons usually rely on physical buttons or visual indicator lights on the device itself. In scenarios where users are moving or their hands are occupied, it is difficult to perform blind operation conveniently, and there is a lack of real-time dynamic data feedback. To address this, this solution proposes a mechanical exoskeleton and safety control system based on human-computer voice interaction. Summary of the Invention
[0005] The purpose of this invention is to provide a mechanical exoskeleton and safety control system based on human-computer voice interaction to solve the technical problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a mechanical exoskeleton based on human-computer voice interaction, comprising an exoskeleton main unit, wherein the exoskeleton main unit includes a waist belt assembly, a flexible chain support assembly, a waist belt tension adjustment assembly, a power supply assembly, an tilt adjustment assembly, and a thigh connection assembly, wherein flexible chain support assemblies are sewn to both sides of the waist belt assembly.
[0007] Furthermore, the waist belt assembly is located on both sides below the flexible chain support assembly and is fixed with tilt adjustment components that fit closely to the hip joint by means of pre-embedded metal connecting pieces and suture reinforcement.
[0008] Furthermore, a power supply component is sewn onto the front waist side of the belt assembly, and a belt tension adjustment component connected to the flexible chain support component is sewn onto the rear waist center of the belt assembly.
[0009] Furthermore, a hip joint drive motor is provided below the tilt adjustment assembly. The output end of the drive motor is rigidly connected to the input shaft of the thigh connection assembly through a universal joint coupling. The power supply assembly is connected to the tilt adjustment assembly and the waist belt tension adjustment assembly through electrical wires.
[0010] A safety control system for a mechanical exoskeleton based on human-computer voice interaction includes a wireless Bluetooth module, an intelligent control center, and a sensor data acquisition module. The user inputs control intentions to the natural language command module through natural language voice, and the wireless Bluetooth module receives language commands output by the natural language command module through the wireless Bluetooth communication protocol.
[0011] Furthermore, the wireless Bluetooth module employs an integrated Bluetooth audio SoC chip.
[0012] Furthermore, the intelligent control center consists of a language interaction module, a TTS feedback synthesis module, and a speech recognition module. The language interaction module receives language commands issued by the user and transmits them to the speech recognition module. The speech recognition module is configured to convert speech into commands that the exoskeleton host can understand based on the edge computing NLU algorithm, and the control commands output by the speech recognition module are forwarded to the exoskeleton host.
[0013] Furthermore, the waist belt tension adjustment component and tilt adjustment component of the exoskeleton host have built-in sensors. These sensors can be torque sensors, angle sensors, and IMU inertial measurement sensors. The exoskeleton host collects data from these sensors in real time through the sensor data acquisition module and packages it into a status data package.
[0014] Furthermore, the sensor data acquisition module is connected to the language interaction module via a status data packet, and the status data packet is transmitted to the language interaction module via an electrical signal.
[0015] The language interaction module inputs the status data packet into the TTS feedback synthesis module. The TTS feedback synthesis module is configured to convert the text information into a speech audio stream. The speech audio stream output by the TTS feedback synthesis module is transmitted to the speech broadcast module for playback.
[0016] Furthermore, the voice signal played by the voice broadcast module is transmitted to the auditory feedback module of the wireless Bluetooth module, and the auditory feedback module transmits the voice broadcast information to the user. The intelligent control center also receives status data packets uplink through the wireless Bluetooth module.
[0017] Compared with the prior art, the beneficial effects of the present invention are:
[0018] 1. This invention issues voice commands through the user's natural language command module, which are then transmitted to the intelligent control center via a wireless Bluetooth module. This design eliminates the limitations of physical buttons or mobile phone screens, and does not require the user to stop or look down to operate. Even when both hands are occupied or in motion, efficient and safe blind operation control can still be achieved through natural language input, greatly improving the ease of operation in complex work scenarios.
[0019] 2. This invention uses a wireless Bluetooth module with an external design that is independent of the exoskeleton host and is worn on the user's collar or chest, achieving physical decoupling between the voice interaction hardware and the mechanical body. This structure effectively avoids direct interference from the vibration of the motor and mechanical friction noise in the tilt adjustment component. Combined with an integrated Bluetooth audio SoC chip, it significantly improves the signal-to-noise ratio and accuracy of voice recognition.
[0020] 3. This invention collects and transmits data in real time through a sensor data acquisition module, and converts the status into voice broadcast through a TTS feedback synthesis module, thereby achieving zero-delay closed-loop control;
[0021] 4. This invention proactively transmits the device status to the user in voice form through the voice broadcast module and auditory feedback module, providing an immersive interactive experience that is independent of visual reliance. In addition, the wireless Bluetooth module has reserved an interface for smartwatches or mobile apps, which can synchronize status data packets to mobile terminals, providing a hardware foundation for health monitoring, data recording and multi-device interconnection ecosystem expansion. Attached Figure Description
[0022] Figure 1 This is a schematic diagram of the main structure of the exoskeleton of the present invention;
[0023] Figure 2 This is a schematic diagram of the wireless Bluetooth module structure of the present invention;
[0024] Figure 3 This is a schematic diagram of the flexible chain support component structure of the present invention;
[0025] Figure 4 This is a schematic diagram of the main frame structure of the mechanical exoskeleton and safety control system based on human-computer voice interaction of the present invention.
[0026] Figure 5 This is a schematic diagram of the framework structure of the voice broadcast module of the present invention;
[0027] Figure 6 This is a schematic diagram of the framework structure of the language interaction module of the present invention.
[0028] In the diagram: 1. Exoskeleton main unit; 101. Waist belt assembly; 102. Flexible chain support assembly; 103. Waist belt tension adjustment assembly; 104. Power supply assembly; 105. Tilt adjustment assembly; 106. Thigh connection assembly; 2. Wireless Bluetooth module; 3. Voice recognition module; 4. Intelligent control center; 5. Language interaction module; 6. TTS feedback synthesis module; 7. Sensor data acquisition module; 8. Auditory feedback module; 9. Natural language command module; 10. Voice broadcasting module. Detailed Implementation
[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0030] Please see Figures 1-5 The present invention provides a technical solution: a mechanical exoskeleton based on human-computer voice interaction, including an exoskeleton host 1. The exoskeleton host 1 includes a waist belt assembly 101, a flexible chain support assembly 102, a waist belt tension adjustment assembly 103, a power supply assembly 104, an angle adjustment assembly 105, and a thigh connection assembly 106. Both sides of the waist belt assembly 101 are sewn with flexible chain support assemblies 102. The waist belt assembly 101 is located below the flexible chain support assembly 102 on both sides and is sewn and fixed with angle adjustment assemblies 105 that fit closely to the hip joint by means of pre-embedded metal connecting pieces and suture reinforcement.
[0031] A power supply component 104 is sewn onto the front waist side of the waist belt assembly 101. A waist belt tension adjustment component 103, connected to the flexible chain support component 102, is sewn onto the center of the back waist of the waist belt assembly 101. A hip joint drive motor is provided below the tilt adjustment component 105. The output end of the motor is rigidly connected to the input shaft of the thigh connection component 106 through a universal joint coupling. The power supply component 104 is connected to the tilt adjustment component 105 and the waist belt tension adjustment component 103 through electrical wires.
[0032] In practice, before use, the user first puts on the belt. The waist belt assembly 101 is adapted to the size by the waist belt tension adjustment assembly 103 at the center of the back waist. The waist belt tension adjustment assembly 103 changes the waist belt circumference according to the user's waist circumference, so that the waist belt assembly 101 fits the user's waist tightly. At the same time, the flexible chain support assembly 102 conforms to the curve of the waist and back, ensuring that the load or assist force can be effectively transmitted to the torso.
[0033] The power assembly 104, sewn to the front waist, not only provides energy, but its front-mounted layout also forms a front-to-back weight balance with the elastic adjustment assembly at the back waist, improving wearing stability. When the system receives an assist command, the hip joint drive motor in the tilt adjustment assembly 105 starts and outputs rotational torque. The power is finally transmitted to the thigh connection assembly 106, which drives the thigh rod to generate assist torque, assisting the user in completing actions such as raising the leg and bending over.
[0034] A safety control system for a mechanical exoskeleton based on human-computer voice interaction includes a wireless Bluetooth module 2, an intelligent control center 4, and a sensor data acquisition module 7.
[0035] See Figures 4-6 It is known that the wireless Bluetooth module 2 receives language commands output by the natural language command module 9 through the wireless Bluetooth communication protocol. The wireless Bluetooth module 2 has reserved an interface for smartwatches or mobile apps. The wireless Bluetooth module 2 adopts an integrated Bluetooth audio SoC chip.
[0036] In practice, within the sound reception range of the natural language command module 9, users issue voice commands such as "mode switch" and "increase power." Through natural language input, users do not need to rely on a mobile phone screen or stop their current action and look down to find physical buttons, which greatly improves the convenience and safety of operation in complex tasks or sports scenarios. The natural language command module 9 collects voice signals, and after preprocessing such as noise reduction and frame segmentation, it sends them to the intelligent control center 4 through the wireless Bluetooth module 2. The wireless Bluetooth module 2 supports two-way communication, realizing downlink transmission of voice commands and uplink reception of status data packets. The wireless Bluetooth module 2 can also transmit status data to a smartwatch or mobile APP via electrical signals.
[0037] See Figures 1-5 It is known that the intelligent control center 4 consists of a language interaction module 5, a TTS feedback synthesis module 6, and a speech recognition module 3. The language interaction module 5 receives language commands issued by the user and transmits them to the speech recognition module 3. The speech recognition module 3 converts speech into commands that the exoskeleton host 1 can understand based on the edge computing NLU algorithm. The control commands output by the speech recognition module 3 are forwarded to the exoskeleton host 1. The motor controller of the exoskeleton host 1 executes the commands and drives the waist belt tension adjustment component 103 and the tilt adjustment component 105 to perform actions.
[0038] In specific implementation, the language interaction module 5 receives the instruction and transmits it to the speech recognition module 3. The speech recognition module 3 uses the edge computing NLU algorithm to perform semantic parsing of the speech and convert it into a digital control instruction that the exoskeleton host 1 can recognize. The instruction is forwarded to the motor controller of the exoskeleton host 1. The controller then drives the adjustment motor in the waist belt tension adjustment component 103 to tighten or loosen, and drives the hip joint drive motor in the tilt adjustment component 105 to output assist torque.
[0039] It effectively avoids the direct interference of motor vibration and mechanical friction noise on the microphone array, significantly improving the signal-to-noise ratio and accuracy of speech recognition.
[0040] See Figures 1-4 It is known that the waist belt tension adjustment component 103 and tilt adjustment component 105 of the exoskeleton host 1 are equipped with sensors. These sensors can be torque sensors, angle sensors and IMU inertial measurement sensors. The exoskeleton host 1 collects the data of these sensors in real time through the sensor data acquisition module 7 and packages it into a status data package.
[0041] In practice, the angle sensor inside the waist belt tension adjustment component 103 is used to detect the tightness of the waist belt, while the tilt adjustment component 105 integrates an IMU inertial measurement sensor and a torque sensor, which are used to monitor the pitch angle of the hip joint and the torque applied by the user's legs, respectively. These sensors collect physical data in real time, and the sensor data acquisition module 7 reads this data at a set frequency and encapsulates it into a structured status data packet according to a predetermined communication protocol.
[0042] See Figures 4-6 It is known that the sensor data acquisition module 7 is connected to the language interaction module 5 through a status data packet. The status data packet is transmitted to the language interaction module 5 through an electrical signal. The language interaction module 5 inputs the status data packet into the TTS feedback synthesis module 6. The TTS module of the TTS feedback synthesis module 6 converts the text information into a speech audio stream. The speech audio stream output by the TTS feedback synthesis module 6 is transmitted to the speech broadcast module 10 and played.
[0043] In practice, the encapsulated status data packet is transmitted to the language interaction module 5 via an electrical signal. The language interaction module 5 extracts the key parameters from the data packet and inputs them into the TTS feedback synthesis module 6. The TTS module calls the built-in text-to-speech engine to convert the text information into a speech audio stream, which is then transmitted to the voice broadcast module 10 to play prompts for "motion status", "motion time", and key device information through a speaker.
[0044] See Figure 2 and Figures 4-6It is known that the voice signal played by the voice broadcast module 10 is transmitted to the auditory feedback module 8 of the wireless Bluetooth module 2. The auditory feedback module 8 transmits the voice broadcast information to the user. The intelligent control center 4 also receives status data packets through the wireless Bluetooth module 2 to ensure real-time synchronization between control commands and device status. The wireless Bluetooth module 2 is worn in the user's collar or chest pocket, achieving overall physical separation from the exoskeleton host 1.
[0045] In practice, the voice signal played by the voice broadcast module 10 is transmitted through the air to the auditory feedback module 8 of the wireless Bluetooth module 2, and finally to the user's ears. At the same time, in order to ensure the consistency of control, the intelligent control center 4 also continuously receives status data packets from the sensor data acquisition module 7 via the wireless Bluetooth module 2. This auditory feedback provides the user with an immersive interactive experience and greatly enhances the perception of human-machine collaboration.
[0046] In summary, when this exoskeleton and safety control system is in operation, the user inputs voice through the natural language command module 9, which is transmitted to the intelligent control center 4 via the wireless Bluetooth module 2. The language interaction module 5 within the center receives the command, and the voice recognition module 3 parses it into digital commands based on the edge computing NLU algorithm. This drives the waist belt tension adjustment component 103 and the thigh connection component 106 of the exoskeleton host 1 to move. The motor of the tilt adjustment component 105 transmits torque through the universal joint coupling. The sensor data acquisition module 7 collects status data and transmits it back. This data is converted into voice by the TTS feedback synthesis module 6 and played by the voice broadcast module 10. The voice feedback is then transmitted to the user through the auditory feedback module 8, thus achieving closed-loop control. The content not described in detail in this specification belongs to the prior art known to those skilled in the art.
[0047] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A mechanical exoskeleton based on human-computer voice interaction, comprising an exoskeleton main unit (1), characterized in that: The exoskeleton main unit (1) includes a waist belt assembly (101), a flexible chain support assembly (102), a waist belt tension adjustment assembly (103), a power supply assembly (104), an angle adjustment assembly (105), and a thigh connection assembly (106). Both sides of the waist belt assembly (101) are sewn with flexible chain support assemblies (102) that fit flexibly.
2. The mechanical exoskeleton based on human-computer voice interaction according to claim 1, characterized in that: The waist belt assembly (101) is located below the flexible chain support assembly (102) on both sides, and is fixed with tilt adjustment components (105) that fit closely to the hip joint by means of pre-embedded metal connecting pieces and suture reinforcement.
3. The mechanical exoskeleton based on human-computer voice interaction according to claim 2, characterized in that: The power supply assembly (104) is sewn to the front waist side of the waist belt assembly (101), and the waist belt tension adjustment assembly (103) connected to the flexible chain support assembly (102) is sewn to the center of the back waist of the waist belt assembly (101).
4. The mechanical exoskeleton based on human-computer voice interaction according to claim 3, characterized in that: The tilt adjustment assembly (105) is provided with a hip joint drive motor below it. The output end of the drive motor is rigidly connected to the input shaft of the thigh connection assembly (106) through a universal joint coupling. The power supply assembly (104) is connected to the tilt adjustment assembly (105) and the waist belt tension adjustment assembly (103) through electrical wires.
5. A safety control system for a mechanical exoskeleton based on human-computer voice interaction, employing the mechanical exoskeleton based on human-computer voice interaction as described in any one of claims 1-4, comprising a wireless Bluetooth module (2), an intelligent control center (4), and a sensor data acquisition module (7), characterized in that: The user inputs control intent to the natural language instruction module (9) through natural language voice, and the wireless Bluetooth module (2) receives the language instructions output by the natural language instruction module (9) through the wireless Bluetooth communication protocol.
6. A safety control system for a mechanical exoskeleton based on human-computer voice interaction according to claim 5, characterized in that: The wireless Bluetooth module (2) adopts an integrated Bluetooth audio SoC chip.
7. A safety control system for a mechanical exoskeleton based on human-computer voice interaction according to claim 6, characterized in that: The intelligent control center (4) consists of a language interaction module (5), a TTS feedback synthesis module (6), and a speech recognition module (3). The language interaction module (5) receives language commands issued by the user and transmits them to the speech recognition module (3). The speech recognition module (3) is configured to convert speech into commands that the exoskeleton host (1) can understand based on the edge computing NLU algorithm, and the control commands output by the speech recognition module (3) are forwarded to the exoskeleton host (1).
8. A safety control system for a mechanical exoskeleton based on human-computer voice interaction according to claim 7, characterized in that: The waist belt tension adjustment component (103) and tilt adjustment component (105) of the exoskeleton host (1) are equipped with sensors. These sensors can be torque sensors, angle sensors and IMU inertial measurement sensors. The exoskeleton host (1) collects the data of these sensors in real time through the sensor data acquisition module (7) and packages it into a status data package.
9. A safety control system for a mechanical exoskeleton based on human-computer voice interaction according to claim 8, characterized in that: The sensor data acquisition module (7) is connected to the language interaction module (5) through a status data packet, and the status data packet is transmitted to the language interaction module (5) through an electrical signal. The language interaction module (5) inputs the status data packet into the TTS feedback synthesis module (6). The TTS feedback synthesis module (6) is configured to convert text information into a speech audio stream. The speech audio stream output by the TTS feedback synthesis module (6) is transmitted to the speech broadcast module (10) and played.
10. A safety control system for a mechanical exoskeleton based on human-computer voice interaction according to claim 9, characterized in that: The voice signal played by the voice broadcast module (10) is transmitted to the auditory feedback module (8) of the wireless Bluetooth module (2). The auditory feedback module (8) transmits the voice broadcast information to the user. The intelligent control center (4) also receives status data packets uplink through the wireless Bluetooth module (2).