Lower limb exoskeleton control weight distribution prediction method based on artificial intelligence

Abstract

The invention discloses a lower limb exoskeleton control weight distribution prediction method based on artificial intelligence. Comprising the following steps that a, biomechanical data, including but not limited to joint angles, muscle activities and ground counter-acting force, of the lower limbs of a wearer are collected through a plurality of sensors; b, a deep learning model is utilized, and the model is trained through a large amount of lower limb movement data in advance to recognize weight distribution characteristics in different movement modes; c, according to the collected biomechanical data, the model predicts the weight distribution borne by each part of the lower limb in real time; and d, according to a prediction result, intelligently adjusting the supporting force of each joint of the exoskeleton so as to optimize the gait of the wearer and reduce energy consumption. Energy consumption and exercise efficiency are optimized, specifically, according to the predicted weight distribution, the supporting strategy of the exoskeleton is intelligently optimized, the energy consumption of the wearer in the exercise process is reduced, and the exercise efficiency is improved. And by reducing unnecessary energy waste, the service life of the battery is prolonged, and the practicability of the exoskeleton is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence. Background Art

[0002] Before the development of lower limb exoskeleton technology, people lacked effective auxiliary equipment to reduce the burden on their bodies and improve exercise efficiency when performing various activities, especially for those with limited mobility, manual laborers, and those who need to perform high-intensity exercise. Traditional auxiliary equipment often has a single function and cannot be accurately adjusted according to the different characteristics of individuals, making it difficult to meet the needs of people of different weights, ages, and physical conditions.

[0003] In terms of energy utilization, there is no intelligent weight distribution prediction and support strategy adjustment mechanism, resulting in high energy consumption and low efficiency during exercise. For example, some mobility assistive devices cannot be dynamically adjusted according to the user's actual movement state, resulting in serious energy waste and limited battery life, which greatly limits the practicality and use time of the device.

[0004] At the same time, existing sports assistive devices usually lack personalized settings and cannot meet the specific needs of different users. When people use these devices, they can only passively adapt to the fixed mode of the device and cannot adjust it according to their physical condition and sports preferences, resulting in a poor user experience and difficulty in achieving truly personalized assistance.

[0005] With the continuous advancement of science and technology, people have higher and higher requirements for sports assistive devices. There is an urgent need for a lower limb exoskeleton technology that can intelligently predict weight distribution, optimize energy consumption, adapt to different groups of people and provide personalized settings to meet the increasingly diverse sports and rehabilitation needs. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the defects of the above-mentioned technology and provide a method for predicting the weight distribution of lower limb exoskeleton control based on artificial intelligence.

[0007] In order to solve the above technical problems, the technical solution provided by the present invention is a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence, which comprises the following steps:

[0008] a. Collect biomechanical data of the wearer's lower limbs through multiple sensors, including but not limited to joint angles, muscle activity and ground reaction force;

[0009] b. Using a deep learning model that is pre-trained with a large amount of lower limb motion data to identify weight distribution characteristics under different motion patterns;

[0010] c. Based on the collected biomechanical data, the model predicts the weight distribution of each part of the lower limb in real time;

[0011] d. Based on the prediction results, intelligently adjust the support force of each joint of the exoskeleton to optimize the wearer's gait and reduce energy consumption.

[0012] As an improvement, the deep learning model is a recurrent neural network (RNN) or a long short-term memory network (LSTM), which can process time series data and improve prediction accuracy.

[0013] As an improvement, the biomechanical data further includes the wearer's heart rate, breathing rate and body temperature to more comprehensively assess the wearer's physical condition.

[0014] As an improvement, the support force adjustment of each joint of the exoskeleton is achieved through a built-in electric motor to ensure fast response and precise control.

[0015] As an improvement, a user interface is also included to display the predicted weight distribution and exoskeleton status, as well as allowing the user to adjust preset movement modes.

[0016] As an improvement, the model training phase also includes generalizing the data of wearers of different weights, ages and physical conditions to enhance the applicability and robustness of the model.

[0017] As an improvement, a feedback mechanism is also included to adjust the model in real time based on the wearer's actual gait and movement feedback to optimize prediction accuracy.

[0018] The advantages of the present invention compared with the prior art are: Optimizing energy consumption and movement efficiency: According to the predicted weight distribution, the support strategy of the exoskeleton is intelligently optimized to reduce the wearer's energy consumption during exercise and improve exercise efficiency.

[0019] By reducing unnecessary energy waste, battery life is extended and the practicality of the exoskeleton is enhanced.

[0020] Enhanced adaptability and personalization: The model is generalized to the data of wearers with different weights, ages, and physical conditions, and can adapt to a wider range of wearers, improving the universal applicability of the exoskeleton.

[0021] The user interface allows the wearer to adjust the preset sports modes according to personal needs, achieving a high degree of personalization.

[0022] Real-time monitoring and health assessment: The integrated heart rate, respiratory rate and body temperature monitoring functions not only support weight distribution prediction, but also comprehensively assess the wearer's physical condition and provide health data reference.

[0023] This data can be used to optimize sports training or provide vital sign monitoring in emergency situations to improve overall safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is an overall schematic diagram of a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence of the present invention.

[0025] Figure 2 It is a schematic diagram of a deep learning model of a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence in the present invention.

[0026] Figure 3 It is a schematic diagram of each joint of an exoskeleton according to an artificial intelligence-based lower limb exoskeleton control weight distribution prediction method of the present invention.

[0027] Figure 4 It is a user interface schematic diagram of a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence of the present invention.

[0028] Figure 5 It is a schematic diagram of the model training phase of a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence of the present invention.

[0029] Figure 6 It is a schematic diagram of the feedback mechanism of a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0031] In the description of the embodiments of the present invention, it should be noted that if the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, or the orientation or position relationship in which the invented product is usually placed when used. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0032] In addition, the terms "horizontal", "vertical", "overhanging" and the like do not mean that the components must be absolutely horizontal or overhanging, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0033] In the description of the embodiments of the present invention, "plurality" means at least 2.

[0034] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0035] In conjunction with the accompanying drawings, a method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence includes the following steps:

[0036] a. Collect biomechanical data of the wearer's lower limbs through multiple sensors, including but not limited to joint angles, muscle activity and ground reaction force;

[0037] b. Using a deep learning model that is pre-trained with a large amount of lower limb motion data to identify weight distribution characteristics under different motion patterns;

[0038] c. Based on the collected biomechanical data, the model predicts the weight distribution of each part of the lower limb in real time;

[0039] d. Based on the prediction results, intelligently adjust the support force of each joint of the exoskeleton to optimize the wearer's gait and reduce energy consumption.

[0040] The deep learning model is a recurrent neural network (RNN) or a long short-term memory network (LSTM), which can process time series data and improve prediction accuracy.

[0041] The biomechanical data further includes the wearer's heart rate, breathing rate and body temperature to more comprehensively assess the wearer's physical condition.

[0042] The support force adjustment of each joint of the exoskeleton is achieved through a built-in electric motor to ensure fast response and precise control.

[0043] Also included is a user interface that displays predicted weight distribution and exoskeleton status, as well as allowing the user to adjust preset motion patterns.

[0044] The model training phase also includes generalizing data of wearers of different weights, ages and physical conditions to enhance the applicability and robustness of the model.

[0045] A feedback mechanism is also included to make real-time adjustments to the model based on the wearer’s actual gait and motion feedback to optimize prediction accuracy.

[0046] 1. Equipment preparation:

[0047] First, make sure the lower limb exoskeleton device is in normal working condition, check whether the various sensors, deep learning models, electric motors, user interfaces, feedback mechanisms and other components are operating normally. At the same time, confirm that the device is connected to the power supply and perform the necessary initialization settings.

[0048] 2. Data Collection Phase:

[0049] Sensor installation and calibration:

[0050] Multiple sensors for collecting biomechanical data are properly installed at the corresponding positions of the wearer's lower limbs. For example, joint angle sensors are installed at major joints such as the knee and hip joints; muscle activity sensors can be fixed near major muscle groups by sticking or strapping; and ground reaction force sensors are installed at the soles of shoes or the parts where the exoskeleton contacts the ground.

[0051] Calibrate each sensor to ensure it is measuring the correct data. This can be done by testing under known conditions, such as measuring ground reaction force under a standard weight, or verifying the accuracy of a joint angle sensor at a specific angle.

[0052] Biomechanical data collection:

[0053] When the wearer starts to move, the sensor begins to collect biomechanical data of the lower limbs. The joint angle sensor monitors the bending and extension angles of the joints in real time and records the joint movement trajectory of the wearer in different motion states.

[0054] The muscle activity sensor detects the contraction and relaxation of muscles and reflects the degree of muscle activity by measuring electromyographic signals.

[0055] Ground reaction force sensors measure the force applied by the wearer to the ground during walking, running or other exercise, and the reaction force fed back to the wearer by the ground.

[0056] At the same time, physiological data such as the wearer's heart rate, breathing rate and body temperature are collected. Heart rate can be measured by a heart rate sensor worn on the wrist or chest; breathing rate can be detected by a sensor on the chest or abdomen; body temperature can be obtained by a contact or non-contact temperature sensor.

[0057] 3. Deep learning model operation stage:

[0058] Model selection and loading:

[0059] Select appropriate deep learning models according to actual needs, such as recurrent neural networks (RNN) or long short-term memory networks (LSTM). These models have the ability to process time series data and can better capture the dynamic characteristics of lower limb movements.

[0060] The pre-trained model is loaded into the exoskeleton's control system. During the training phase, the model has been generalized using a large amount of lower limb motion data from wearers of different weights, ages, and physical conditions to enhance its applicability and robustness.

[0061] Data input and prediction:

[0062] The collected biomechanical data, including joint angles, muscle activity, ground reaction forces, heart rate, breathing rate, and body temperature, are fed into the deep learning model.

[0063] Based on the input data, the model identifies the weight distribution characteristics in different movement modes and predicts the weight distribution of each part of the lower limb in real time. For example, during walking, the model can predict the weight proportion borne by the knee, hip and ankle joints.

[0064] 4. Exoskeleton support force adjustment stage:

[0065] Electric Motor Control:

[0066] The electric motors built into each joint of the exoskeleton adjust the support force based on the weight distribution results predicted by the model. If a joint bears a large amount of weight, the electric motor will increase the support force of the joint to reduce the burden on the wearer.

[0067] The electric motor adjustment is fast-response and precisely controlled. Through the control system algorithm, the motor can dynamically adjust according to the real-time weight distribution changes, ensuring that the wearer can get the best support in different motion states.

[0068] Optimized gait and reduced energy expenditure:

[0069] The adjusted support force helps optimize the wearer's gait. For example, by properly distributing the support force of each joint, the wearer's steps can be made more stable and natural, reducing the extra energy consumption caused by imbalance.

[0070] The exoskeleton's control system can automatically adjust the support force to achieve the best gait optimization effect according to the wearer's sports goals and physical condition. For example, when running, increasing the support force of the legs can improve running efficiency; when going up and down stairs, adjusting the angle and support force of the joints can make the movement easier.

[0071] 5. User interface and feedback mechanism:

[0072] The user interface displays:

[0073] The user interface can display the predicted weight distribution and the status information of the exoskeleton. For example, it can graphically display the weight proportion of each part of the lower limbs, as well as the current support force and movement speed of each joint of the exoskeleton.

[0074] Users can understand their movement status and the working status of the exoskeleton through the user interface so as to adjust their movement method or seek help in time.

[0075] User-adjustable preset sport modes:

[0076] The user interface allows users to adjust preset motion modes. For example, users can select different motion modes, such as walking, running, going up and down stairs, etc. The exoskeleton will automatically adjust the support force and motion parameters according to the mode selected by the user.

[0077] Users can also customize the parameters of the exercise mode, such as step length, speed, support force, etc. according to their physical condition and needs.

[0078] Feedback mechanism operation:

[0079] The feedback mechanism adjusts the model in real time based on the wearer’s actual gait and motion feedback. For example, when the wearer’s gait changes, the sensor detects new biomechanical data, and the feedback mechanism feeds this data back to the deep learning model.

[0080] The model makes real-time adjustments based on the feedback data to optimize the prediction accuracy. At the same time, the exoskeleton's control system will adjust the support force again based on the adjusted prediction results to ensure that the wearer always gets the best sports assistance effect.

[0081] Optimize energy consumption and exercise efficiency: Based on the predicted weight distribution, the exoskeleton's support strategy is intelligently optimized to reduce the wearer's energy consumption during exercise and improve exercise efficiency.

[0082] By reducing unnecessary energy waste, battery life is extended and the practicality of the exoskeleton is enhanced.

[0083] Enhanced adaptability and personalization: The model is generalized to the data of wearers with different weights, ages, and physical conditions, and can adapt to a wider range of wearers, improving the universal applicability of the exoskeleton.

[0084] The user interface allows the wearer to adjust the preset sports modes according to personal needs, achieving a high degree of personalization.

[0085] Real-time monitoring and health assessment: The integrated heart rate, respiratory rate and body temperature monitoring functions not only support weight distribution prediction, but also comprehensively assess the wearer's physical condition and provide health data reference.

[0086] This data can be used to optimize sports training or provide vital sign monitoring in emergency situations to improve overall safety.

[0087] Continuous optimization and feedback: The feedback mechanism ensures that the system can adjust in real time according to the wearer's actual sports experience, continuously optimize the model's prediction accuracy, and improve overall system performance.

[0088] Through continuous data collection and model iteration, the system can gradually adapt to more complex sports scenarios and provide more accurate and intelligent services.

[0089] Improve wearing comfort and safety: By accurately predicting the weight distribution of each part of the lower limbs and intelligently adjusting the exoskeleton support force, the wearer's comfort can be significantly improved and muscle fatigue or injury caused by uneven load can be reduced.

[0090] Real-time feedback and adjustment mechanism ensures that the exoskeleton can provide optimal support and improve exercise safety and stability during the wearer's exercise.

[0091] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence, characterized in that: The following steps are involved: a. Collect biomechanical data of the wearer's lower limbs through multiple sensors, including but not limited to joint angles, muscle activity and ground reaction force; b. Using a deep learning model that is pre-trained with a large amount of lower limb motion data to identify weight distribution characteristics under different motion patterns; c. Based on the collected biomechanical data, the model predicts the weight distribution of each part of the lower limb in real time; d. Based on the prediction results, intelligently adjust the support force of each joint of the exoskeleton to optimize the wearer's gait and reduce energy consumption.

2. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 1, characterized in that: The deep learning model is a recurrent neural network (RNN) or a long short-term memory network (LSTM), which can process time series data and improve prediction accuracy.

3. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 2, characterized in that: The biomechanical data further includes the wearer's heart rate, breathing rate and body temperature to more comprehensively assess the wearer's physical condition.

4. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 3, characterized in that: The support force adjustment of each joint of the exoskeleton is achieved through a built-in electric motor to ensure fast response and precise control.

5. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 4, characterized in that: Also included is a user interface that displays predicted weight distribution and exoskeleton status, as well as allowing the user to adjust preset motion patterns.

6. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 5, characterized in that: The model training phase also includes generalizing data of wearers of different weights, ages and physical conditions to enhance the applicability and robustness of the model.

7. The method for predicting weight distribution of lower limb exoskeleton control based on artificial intelligence according to claim 6, characterized in that: A feedback mechanism is also included to make real-time adjustments to the model based on the wearer’s actual gait and motion feedback to optimize prediction accuracy.

Images