Intelligent cabin control system and control method based on multi-source perception and adaptive learning

Abstract

The invention belongs to the field of intelligent automobile cabin control, and particularly relates to an intelligent cabin control system and control method based on multi-source sensing and adaptive learning, and the system comprises a multi-source sensing module, a user interaction module, a data processing module, an adaptive learning module, a control module and an execution module. The multi-source sensing module collects information and inputs the information into the data processing module, the data processing module processes the received information and inputs the information into the self-adaptive learning module, and the self-adaptive learning module inputs the information input by the data processing module into a machine learning model by adopting a machine learning technology to obtain a prediction result. A control instruction is output to the control module according to the prediction result, and the control module controls the execution module to execute the control instruction. The environment of the cabin can be automatically adjusted according to user requirements, and personalized services are provided for users.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent automobile cockpit control, and in particular, relates to an intelligent cockpit control system and a control method based on multi-source perception and adaptive learning. Background Art

[0002] With the continuous improvement of the level of automobile intelligence, the intelligent cockpit, as the core part of the interior of the car, has a direct impact on the user's driving and riding experience. Chinese Patent 219277292U relates to an intelligent driving cockpit and vehicle. The intelligent driving cockpit includes an image sensor arranged in the driving cockpit, a controller electrically connected to the image sensor, an electric adjustment seat arranged in the driving cockpit and electrically connected to the controller, and a communicator. The communicator is electrically connected to the controller and the ECU, and is used to obtain the accelerator pedal tilt data and the brake pedal tilt data through the ECU. The controller is used to adjust the initial position of the electric adjustment seat according to the feedback of the image sensor. This patent achieves the purpose of adjusting the seat through sensor feedback, but relies on set rules to adjust the cockpit environment, and cannot achieve precise environmental adjustment and personalized services. Summary of the invention

[0003] The present invention aims to provide an intelligent cockpit control system and control method based on multi-source perception and adaptive learning, so as to achieve the technical purpose of improving the intelligent adjustment function of the cabin environment and user experience through multi-source perception technology and adaptive learning algorithm.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] The present invention provides an intelligent cockpit control system based on multi-source perception and adaptive learning, comprising a multi-source perception module, a user interaction module, a data processing module, an adaptive learning module, a control module and an execution module, wherein the output ends of the multi-source perception module and the user interaction module are respectively connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the adaptive learning module, the output end of the adaptive learning module is connected to the input end of the control module, and the output end of the control module is connected to the input end of the execution module.

[0006] The multi-source perception module and the user interaction module input the collected information to the adaptive learning module through the data processing module, and the adaptive learning module outputs control instructions to the execution module through the control module.

[0007] The multi-source perception module includes sensors and perception devices, the control module includes a cockpit domain control module and a body domain control module, and the execution module includes an air-conditioning module, a power amplifier module, an ambient light module and a seat module; the sensors and perception devices are respectively connected to the data processing module; the output end of the cockpit domain control module is respectively connected to the input end of the air-conditioning module, the power amplifier module and the ambient light module, and the output end of the body domain control module is connected to the input end of the seat module.

[0008] The present invention provides a control method for an intelligent cockpit control system based on multi-source perception and adaptive learning.

[0009] Step 1: The multi-source perception module collects information and inputs it into the data processing module;

[0010] Step 2: The data processing module processes the received information and inputs it into the adaptive learning module;

[0011] Step 3: The adaptive learning module uses machine learning technology to input the information input by the data processing module into the machine learning model to obtain the prediction result, and outputs the control instruction to the control module according to the prediction result;

[0012] Step 4: The control module controls the execution module to execute the control instruction.

[0013] The user inputs information to the data processing module through voice interaction, touch interaction and gesture recognition in the user interaction module.

[0014] In step one, the multi-source perception module includes sensors and perception devices. The multi-source perception module collects environmental information, user's movements and facial expression information through the sensors and perception devices and inputs them into the data processing module.

[0015] In step 2, the data processing module processes the data input from the multi-source perception module and the user interaction module through data preprocessing and data fusion technology, and then inputs the processed data into the adaptive learning module.

[0016] The data processing module cleans, denoises and standardizes the data input from the multi-source perception module and the user interaction module, then extracts features, associates the data through features, fuses the associated data to obtain a consistent description, and inputs the results into the adaptive learning module.

[0017] The machine learning model in the adaptive learning module in step three is obtained by training the machine learning algorithm and historical data, and the machine learning model is repeatedly trained through the machine learning algorithm based on the data input by the data processing module.

[0018] In step 4, the control module is divided into a cockpit domain control module and a body domain control module. The cockpit domain control module outputs control instructions to the air conditioning module, power amplifier module and ambient light module in the execution module, and the body domain control module outputs control instructions to the seat module in the execution module.

[0019] The technical effects of the present invention are:

[0020] (1) The system can sense the environmental information inside and outside the cabin in real time through the multi-source perception module to ensure timely adjustment.

[0021] (2) The user interaction module provides a variety of interaction methods. When users are dissatisfied with the system environment or operation, they can express their needs and preferences through the user interaction module.

[0022] (3) The data processing module receives information input from the multi-source perception module and the user interaction module, cleans, denoises and standardizes the data to improve its accuracy and reliability.

[0023] (4) The adaptive learning module identifies the user's expression needs based on the environmental information and user interaction information input by the data processing module, and provides personalized services to the user through the machine learning model to improve the user experience.

[0024] (5) Through the adaptive learning module, the system can automatically adjust the cabin environment parameters according to historical data and environmental changes, achieving precise intelligent adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] This specification includes the following drawings, which show the following contents:

[0026] Figure 1 The logical structure block diagram of an intelligent cockpit control system and control method based on multi-source perception and adaptive learning of the present invention;

[0027] Figure 1 Marked as: 1. Multi-source perception module; 2. User interaction module; 3. Data processing module; 4. Adaptive learning module; 5. Control module; 6. Execution module. DETAILED DESCRIPTION

[0028] The specific implementation methods of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, with the aim of helping those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention and facilitating their implementation.

[0029] The present invention provides an intelligent cockpit control system based on multi-source perception and adaptive learning, comprising a multi-source perception module 1, a user interaction module 2, a data processing module 3, an adaptive learning module 4, a control module 5 and an execution module 6, wherein the output ends of the multi-source perception module 1 and the user interaction module 2 are respectively connected to the input ends of the data processing module 3, the output end of the data processing module 3 is connected to the input end of the adaptive learning module 4, the output end of the adaptive learning module 4 is connected to the input end of the control module 5, and the output end of the control module 5 is connected to the input end of the execution module 6.

[0030] The multi-source perception module 1 and the user interaction module 2 input the collected information to the adaptive learning module 4 through the data processing module 3 , and the adaptive learning module 4 outputs control instructions to the execution module 6 through the control module 5 .

[0031] The control module 5 includes a cockpit domain control module and a body domain control module; the execution module 6 includes an air-conditioning module, a power amplifier module, an ambient light module and a seat module; the output end of the cockpit domain control module is respectively connected to the input end of the air-conditioning module, the power amplifier module and the ambient light module, and the output end of the body domain control module is connected to the input end of the seat module.

[0032] The present invention provides a control method for an intelligent cockpit control system based on multi-source perception and adaptive learning, including step one: a multi-source perception module 1 collects information and inputs the information into a data processing module 3; step two: the data processing module 3 processes the received information and inputs the information into an adaptive learning module 4; step three: the adaptive learning module 4 adopts machine learning technology, inputs the information input by the data processing module 3 into a machine learning model to obtain a prediction result, and outputs a control instruction to a control module 5 according to the prediction result; step four: the control module 5 controls the execution module 6 to execute the control instruction.

[0033] The user inputs information to the data processing module 3 through voice interaction, touch interaction and gesture recognition in the user interaction module 2.

[0034] In step 1, the multi-source perception module 1 includes sensors and perception devices. The multi-source perception module 1 collects environmental information, user's movements and facial expression information through sensors and perception devices and inputs them into the data processing module 3. In step 2, the data processing module 3 processes the data input by the multi-source perception module 1 and the user interaction module 2 through data preprocessing and data fusion technology, and then inputs the processed data into the adaptive learning module 4. The data processing module 3 cleans, denoises and standardizes the data input by the multi-source perception module 1 and the user interaction module 2, and then extracts features, associates the data through features, fuses the associated data to obtain a consistent description, and inputs the results into the adaptive learning module 4. In step 3, the machine learning model in the adaptive learning module 4 is obtained by training with a machine learning algorithm and historical data, and the machine learning model is repeatedly trained by a machine learning algorithm according to the data input by the data processing module 3. In step 4, the control module 5 is divided into a cockpit domain control module and a body domain control module. The cockpit domain control module outputs control instructions to the air conditioning module, power amplifier module and ambient light module in the execution module 6, and the body domain control module outputs control instructions to the seat module in the execution module 6.

[0035] The system and control method of the present invention are described in detail below.

[0036] The present invention provides an intelligent cockpit control system based on multi-source perception and adaptive learning, comprising a multi-source perception module 1, a user interaction module 2, a data processing module 3, an adaptive learning module 4, a control module 5 and an execution module 6, wherein the multi-source perception module 1 comprises a sensor and a perception device, the sensor comprises a temperature sensor, a humidity sensor, a light sensor and an air quality sensor, the perception device comprises a facial recognition camera and a radar, and the temperature sensor, the humidity sensor, the light sensor, the air quality sensor, the facial recognition camera and the radar are respectively connected to the data processing module 3. The user interaction module 2 comprises voice interaction, touch interaction and gesture recognition, and adopts a microphone, a touch screen and a gesture recognition camera, which are respectively connected to the data processing module 3. The output end of the data processing module 3 is connected to the input end of the adaptive learning module 4. The control module 5 is divided into a cockpit domain control module and a body domain control module, and the output end of the adaptive learning module 4 is respectively connected to the input end of the cockpit domain control module and the body domain control module. The execution module 6 includes an air conditioning module, a power amplifier module, an ambient light module and a seat module. The output end of the cockpit domain control module is connected to the input end of the air conditioning module, the power amplifier module and the ambient light module, and the output end of the body domain control module is connected to the input end of the seat module. The air conditioning module uses an air conditioner, the power amplifier module uses an amplifier and an equalizer, the ambient light module uses an ambient light, and the seat module uses a seat.

[0037] The multi-source perception module 1 and the data processing module 3 can be connected via the I2C bus, SPI bus, LVDS bus, CAN bus or Ethernet, the user interaction module 2 and the data processing module 3 can be connected via the A2B bus or LVDS bus, the data processing module 3 and the adaptive learning module 4, the adaptive learning module 4 and the control module 5, and the control module 5 and the execution module 6 are connected via the CAN bus or Ethernet. The user interaction module 2, the data processing module 3, the adaptive learning module 4 and the cockpit domain control module use the system-level chip SOC as the main control chip and use the Android or QNX system as the control system. The cockpit domain control module controls the air conditioner, power amplifier, equalizer and atmosphere light through the system-level chip SOC. The body domain control module uses the microcontroller MCU as the control chip, and the body domain control module controls the seat through the microcontroller MCU.

[0038] The following is a detailed description of a control method for an intelligent cockpit control system based on multi-source perception and adaptive learning of the present invention. The multi-source perception module 1 of the present invention includes a temperature sensor, a humidity sensor, a light sensor, an air quality sensor, a facial recognition camera and a radar, which respectively sense the temperature, humidity, light, air quality information inside and outside the cabin, as well as the movements and facial expressions of the passengers, and input these information into the data processing module 3. The user interaction module 2 includes a microphone, a touch screen and a gesture recognition camera, which inputs voice, touch and gesture information into the data processing module 3 through voice interaction, touch interaction and gesture recognition, and the user expresses his wishes through the user interaction module 2. The system can sense the environmental information inside and outside the cabin in real time through the multi-source perception module 1 to ensure the timeliness of the adjustment. When the user is dissatisfied with the system environment or operation, he can express his needs and preferences through the user interaction module 2.

[0039] The data processing module 3 receives the information input by the multi-source perception module 1 and the user interaction module 2, cleans, denoises and standardizes the data to improve its accuracy and reliability, and then performs data fusion, that is, collects information data of the environment or user through multiple different types of sensors, performs feature extraction transformation on these data, extracts feature vectors; performs recognition processing on the feature vectors, completes the description of the environment or user by each sensor and perception device; groups the collected information data of the environment or user through the description, that is, associates, and synthesizes the various data of the environment or user using the fusion algorithm to obtain information data with consistent interpretation and description of the environment or user, thereby improving the accuracy of the information. The data processing module 3 inputs the data after data preprocessing and data fusion into the adaptive learning module 4. The data processing module 3 can accurately identify the environmental conditions and the user's emotions and intentions through sensors and perception devices, and prepare for subsequent adjustments.

[0040] The adaptive learning module 4 uses machine learning technology to input the environmental information and user facial expression or action information input by the data processing module 3 into the machine learning model to obtain the prediction result, judge the environmental situation or the user's emotion, and output the control instruction to the control module 5 according to the environmental situation or the user's expression; the machine learning model in the adaptive learning module 4 is obtained by training the machine learning algorithm and historical data, and the machine learning model is repeatedly trained by the machine learning algorithm according to the data input by the data processing module 3. Among them, the historical data is universal, but in order to provide personalized services according to user needs, the machine learning model is repeatedly trained according to the data input by the data processing module 3; for example, if the user likes lower temperature and humidity, the machine learning model will be trained to provide a model with lower temperature and humidity. The adaptive learning module 4 identifies the user's expression needs based on the environmental information and user interaction information input by the data processing module 3, and provides personalized services to the user through the machine learning model.

[0041] The adaptive learning module 4 sends control instructions to the control module 5 to control the work of the execution module 6 according to the results obtained by the machine learning model. In the embodiment of the present invention, the control module 5 controls the air conditioning module, the power amplifier module, the atmosphere light module and the seat module, that is, controls the air conditioning to control the temperature and humidity of the environment, controls the power amplifier and the equalizer in the power amplifier module to control the sound switch and the volume and quality of the sound, controls the atmosphere light to control the switch of different lights, the brightness and color of the lights, and controls the seat module to control the height, front and back and massage adjustment of the seat. The work of the air conditioning module, the power amplifier module, the atmosphere light module and the seat module is affected and controlled by the user; for example, when the system detects that the user is in a low mood, the system controls the atmosphere light module to emit a soft light with warm tones, and plays soothing music through the power amplifier module to adjust the user's mood.

[0042] The working process of the present invention is described in detail below. The system provides a multi-source perception module 1 to detect environmental information and the user's actions and facial expressions and input them into a data processing module 3. The data processing module 3 obtains more accurate data information through data preprocessing and data fusion, and inputs this information into an adaptive learning module 4. The adaptive learning module 4 inputs this information into the established machine learning model, outputs control instructions to the control module 5 according to the prediction results, and controls the execution module 6 to execute the command. The adaptive learning module 4 will judge the user's emotions based on the collected facial expression information, and issue corresponding instructions based on the emotions, such as playing music when the user is depressed. When the user is dissatisfied with the current state, the user can express the demand through the user interaction module 2, and the demand can be expressed specifically through voice, touch screen and gestures. These data will be input into the adaptive learning module 4 through the data processing module 3. The adaptive learning module 4 records these data, and trains the machine learning model through a machine learning algorithm, so as to make a satisfactory response to the user when a similar situation occurs next time, and provide personalized services for the user.

[0043] The present invention is described above by way of example in conjunction with the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention; or the above concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. An intelligent cockpit control system based on multi-source perception and adaptive learning, Features: It includes a multi-source perception module, a user interaction module, a data processing module, an adaptive learning module, a control module and an execution module. The output ends of the multi-source perception module and the user interaction module are respectively connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the adaptive learning module, the output end of the adaptive learning module is connected to the input end of the control module, and the output end of the control module is connected to the input end of the execution module.

2. An intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 1, Features: The multi-source perception module and the user interaction module input the collected information to the adaptive learning module through the data processing module, and the adaptive learning module outputs control instructions to the execution module through the control module.

3. An intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 1, Features: The multi-source perception module includes sensors and perception devices, the control module includes a cockpit domain control module and a body domain control module, and the execution module includes an air-conditioning module, a power amplifier module, an ambient light module and a seat module; the sensors and perception devices are respectively connected to the data processing module; the output end of the cockpit domain control module is respectively connected to the input end of the air-conditioning module, the power amplifier module and the ambient light module, and the output end of the body domain control module is connected to the input end of the seat module.

4. A control method for an intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in any one of claims 1 to 3, Features: Step 1: The multi-source perception module collects information and inputs it into the data processing module; Step 2: The data processing module processes the received information and inputs it into the adaptive learning module; Step 3: The adaptive learning module uses machine learning technology to input the information input by the data processing module into the machine learning model to obtain the prediction result, and outputs the control instruction to the control module according to the prediction result; Step 4: The control module controls the execution module to execute the control instruction.

5. A control method for an intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, Features: The user inputs information to the data processing module through voice interaction, touch interaction and gesture recognition in the user interaction module.

6. A control method for an intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, Features: In step 1, the multi-source perception module collects environmental information, user's movements and facial expression information through sensors and perception devices and inputs them into the data processing module.

7. The control method of the intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, Features: In step 2, the data processing module processes the data input from the multi-source perception module and the user interaction module through data preprocessing and data fusion technology, and then inputs the processed data into the adaptive learning module.

8. A control method for an intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, 5 or 7, Features: The data processing module cleans, denoises and standardizes the data input from the multi-source perception module and the user interaction module, then extracts features, associates the data through features, fuses the associated data to obtain a consistent description, and inputs the results into the adaptive learning module.

9. The control method of the intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, Features: The machine learning model in the adaptive learning module in step three is obtained by training the machine learning algorithm and historical data, and the machine learning model is repeatedly trained through the machine learning algorithm based on the data input by the data processing module.

10. The control method of the intelligent cockpit control system based on multi-source perception and adaptive learning as claimed in claim 4, Features: In step 4, the cockpit domain control module outputs control instructions to the air conditioning module, power amplifier module and ambient light module in the execution module, and the body domain control module outputs control instructions to the seat module in the execution module.

Images