Intelligent driving mode switching method and system based on automatic driving

Abstract

The invention provides an intelligent driving mode switching method based on automatic driving. The method comprises the following steps: step 1, continuously detecting whether a preset L2 automatic driving mode requirement is met or not; 2, if the requirement of the L2 automatic driving mode is met, whether a chassis, a vehicle body and a cabin have faults or not is judged, if no faults exist, the step 3 is executed, and if faults exist, the step 4 is executed; step 3, sending an L3 automatic driving request to a human-computer interaction interface, and turning to step 7; 4, the intelligent driving system judges and distinguishes the fault condition, if the fault is judged to be a serious fault, the step 5 is executed, and if the fault is judged to be a slight fault, the step 6 is executed; 5, a manual takeover request is sent to the human-computer interaction interface, if no one responds to the takeover request, an L3 automatic driving mode is switched, and a preset minimum risk strategy is executed; step 6, sending an L2 automatic driving request to the human-computer interaction interface, and turning to step 7; and 7, starting a corresponding automatic driving mode.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to an intelligent driving mode switching method and system based on automatic driving. Background Art

[0002] Autonomous driving technology has developed rapidly in recent years, with 2026 being the first year for Level 3 advanced autonomous driving. The intelligent driving capabilities of vehicles with Level 3 advanced autonomous driving will encompass both Level 3 and Level 2 intelligent driving. The Operational Design Domain (ODD) (Operational Design Domain) for Level 3 intelligent driving differs from that for Level 2 intelligent driving. Therefore, the activation / exit of intelligent driving functions must be differentiated between Level 3 and Level 2 intelligent driving. Furthermore, transitions between L3 and L2 intelligent driving modes are permitted when appropriate conditions are met and the driver responds correctly to system requests. However, existing L3 intelligent driving technology does not consider transitions between high- and low-level intelligent driving modes. Summary of the Invention

[0003] In view of this, the present invention provides a method for switching intelligent driving modes based on automatic driving. This technical solution solves the technical problems that the existing technology is difficult to ensure that L3 vehicles can smoothly activate / exit the intelligent driving function, and at the same time cannot achieve mutual switching between high-level and low-level intelligent driving functions, thereby improving the user's automatic driving experience.

[0004] The present invention provides an intelligent driving mode switching method based on automatic driving, which is applied to L3 level automatic driving vehicles. The method includes: step 1, after the vehicle is powered on, continuously detecting whether the vehicle status, road conditions, weather environment, road facilities, target conditions and driver status meet the preset L2 automatic driving mode requirements; step 2, if the L2 automatic driving mode requirements are met, based on the changes in the status of the chassis, body and cabin opponent parts monitoring, it is determined whether the chassis, body and cabin have faults. If there are no faults, the L3 automatic driving mode requirements are met, and step 3 is performed. If there are faults, the fault condition is fed back to the vehicle's intelligent driving system, and step 4 is performed; step 3, an L3 automatic driving request is sent To the human-computer interaction interface, to prompt the driver to switch to L3 autonomous driving mode, go to step 7; In step 4, the intelligent driving system judges and distinguishes the fault situation. If it is judged to be a serious fault, go to step 5; if it is judged to be a minor fault, go to step 6; In step 5, a manual takeover request is sent to the human-computer interaction interface. If no one responds to the takeover request, switch to L3 autonomous driving mode and execute the preset minimum risk strategy; In step 6, an L2 autonomous driving request is sent to the human-computer interaction interface to prompt the driver to switch to L2 autonomous driving mode, go to step 7; In step 7, after issuing the L2 / L3 autonomous driving request, if a start signal issued by the driver is detected, the corresponding autonomous driving mode is started.

[0005] Furthermore, the method also includes: step 8, when the vehicle receives an exit signal sent by the driver, exiting the automatic driving system.

[0006] Furthermore, the driver sends the exit signal specifically when the driver touches a designated button, or when the driver's intervention in the vehicle steering control exceeds a preset threshold, or when the driver steps on the brake pedal.

[0007] Furthermore, step 1 also includes: determining whether the current position of the vehicle is within a preset geographic fence, and if so, monitoring whether the preset automatic driving mode requirements are met.

[0008] Furthermore, step 2 also includes: if the L2 autonomous driving mode requirements are not met, the autonomous driving mode cannot be started.

[0009] Furthermore, step 7 also includes: if no start signal is detected, continuing to send the L2 / L3 autonomous driving request until the driver is detected to respond to the L2 / L3 autonomous driving request, or stopping sending the L2 autonomous driving request after the request continues for a preset time period.

[0010] Furthermore, detecting that the driver responds to the L2 / L3 autonomous driving request includes: detecting that the driver sends a start signal or an exit signal.

[0011] Furthermore, the weather environment detection includes rainfall detection, and the rainfall detection method includes: Step 11, collecting rainfall video and intercepting a fixed number of frames as input data; Step 12: Pixels are grouped according to their similarity in pixel features. The color image is converted into a 5-dimensional feature vector through linear iterative clustering. A distance metric is constructed for the vector, and pixels are locally clustered to generate a superpixel block map. Step 13: Select the middle frame as the key frame and use it as the reference position frame, and form a group of partitioned regions with the superpixel block images of the two frames before and after it; Step 14: construct a loss function to measure the difference between each superpixel block in the superpixel block map and its neighboring blocks in the two previous and next frames, and use the neighboring block with the smallest difference as the candidate alignment area; Step 15: Compare the pixel differences between the superpixel block and the candidate alignment region. If the difference between a pixel in the superpixel block and a pixel corresponding to the candidate alignment region is greater than a threshold, the pixel is determined to be a rain pixel. This judgment is repeated for each pixel in the video to obtain a rain segmentation result. In step 16, the rainwater segmentation results are processed through a deep learning neural network to obtain a rainfall classification result, thereby completing rainfall detection.

[0012] The present invention also provides an intelligent driving mode switching system based on automatic driving, the system comprising: a detection module for continuously detecting whether the vehicle status, road conditions, weather environment, road facilities, target conditions and driver status meet the preset L2 automatic driving mode requirements after the vehicle is powered on; if the L2 automatic driving mode requirements are met, the chassis, body and cockpit are judged based on the changes in the status of the opponent parts monitoring the chassis, body and cockpit to determine whether there are faults, and the fault situation is fed back to the vehicle's intelligent driving system; the intelligent driving system is connected to the detection module, and if there is no fault, an L3 automatic driving request is sent to the human-computer interaction interface; if there is a fault, the fault situation is judged Distinguish: if it is judged to be a serious fault, a manual takeover request is sent to the human-computer interaction interface. If no one responds to the takeover request, it switches to L3 autonomous driving mode and executes the preset minimum risk strategy. If it is judged to be a minor fault, an L2 autonomous driving request is sent to the human-computer interaction interface; the start / exit module is connected to the intelligent driving system and is used to start the corresponding autonomous driving mode if a start signal issued by the driver is detected after the L2 / L3 autonomous driving request is issued; the human-computer interaction interface is connected to the intelligent driving system and is used to display L2 / L3 autonomous driving requests or manual takeover requests to prompt the driver to switch to L2 / L3 autonomous driving mode or manual driving mode.

[0013] Furthermore, the start / exit module is also used to exit the automatic driving system when the vehicle receives an exit signal from the driver.

[0014] This invention provides a method for switching intelligent driving modes based on autonomous driving. It primarily explains how to activate and exit an L3 autonomous driving system and switch between high- and low-level intelligent driving modes. This method ensures that L3 vehicles can smoothly activate and exit intelligent driving functions, while also enabling switching between high- and low-level intelligent driving functions, thereby enhancing the user's autonomous driving experience. This method addresses the existing issues of failing to monitor the driver's ability to take over the vehicle, failing to control the exit of L3 autonomous driving, and failing to consider switching between high- and low-level intelligent driving modes. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for switching intelligent driving modes based on autonomous driving provided by the present invention; Figure 2 This is another flow chart of an intelligent driving mode switching method based on autonomous driving provided by the present invention; Figure 3 This is a flow chart of a processing method after issuing an L2 / L3 autonomous driving request provided by the present invention; Figure 4 This is a flow chart of a rainfall detection method provided by the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] Example 1: The present invention provides a method and system for switching intelligent driving modes based on autonomous driving. The system includes a detection module, an intelligent driving system, a start / exit module, and a human-computer interaction interface, and is applied to L3 autonomous driving vehicles, such as Figure 1 As shown, the method includes the following steps.

[0018] Step 1: After the vehicle is powered on, it continuously checks whether the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status meet the preset L2 autonomous driving mode requirements; Step 2: If the requirements for L2 autonomous driving mode are met, the chassis, body, and cabin components are monitored for changes in state to determine whether there are any faults. If there are no faults, the requirements for L3 autonomous driving mode are met, and the process goes to Step 3. If there are faults, the fault condition is fed back to the vehicle's intelligent driving system, and the process goes to Step 4. The detection module is used to continuously monitor the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status after the vehicle is powered on to determine whether they meet the preset L2 autonomous driving mode requirements. If so, the module determines whether there are any faults in the chassis, body, and cabin based on changes in the status of the opponent components, and reports the faults to the vehicle's intelligent driving system. Step 3: Send an L3 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to L3 autonomous driving mode, and then go to step 7; Step 4: The intelligent driving system determines the fault condition. If it is a serious fault, go to step 5; if it is a minor fault, go to step 6. Step 5: Send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, switch to L3 autonomous driving mode and execute the preset minimum risk strategy; Step 6: Send an L2 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to the L2 autonomous driving mode, and then go to step 7; The intelligent driving system is connected to the detection module. If there is no fault, it will send an L3 autonomous driving request to the human-computer interaction interface. If there is a fault, it will judge and distinguish the fault situation. If it is judged to be a serious fault, it will send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, it will switch to the L3 autonomous driving mode and execute the preset minimum risk strategy. If it is judged to be a minor fault, it will send an L2 autonomous driving request to the human-computer interaction interface; the human-computer interaction interface is connected to the intelligent driving system and is used to display L2 / L3 autonomous driving requests or manual takeover requests to prompt the driver to switch to L2 / L3 autonomous driving mode or manual driving mode.

[0019] Step 7: After issuing an L2 / L3 autonomous driving request, if a start signal issued by the driver is detected, the corresponding autonomous driving mode is started.

[0020] The start / exit module is connected to the intelligent driving system and is used to activate the corresponding autonomous driving mode if a start signal from the driver is detected after an L2 / L3 autonomous driving request is issued; This invention provides a method for switching intelligent driving modes based on autonomous driving. It primarily explains how to activate and exit an L3 autonomous driving system and switch between high- and low-level intelligent driving modes. This method ensures that L3 vehicles can smoothly activate and exit intelligent driving functions, while also enabling switching between high- and low-level intelligent driving functions, thereby enhancing the user's autonomous driving experience. This method addresses the existing issues of failing to monitor the driver's ability to take over the vehicle, failing to control the exit of L3 autonomous driving, and failing to consider switching between high- and low-level intelligent driving modes.

[0021] Example 2: The present invention provides a method and system for switching intelligent driving modes based on autonomous driving, which is applied to L3 autonomous driving vehicles, such as Figure 2 As shown, the method includes the following steps.

[0022] Step 1: After the vehicle is powered on, it continuously checks whether the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status meet the preset L2 autonomous driving mode requirements; After the vehicle's power-on self-test is completed, the system will continuously check whether all L3 ODD and L2 ODD conditions are met. When all L3 ODD or L2 ODD conditions are met, the system should notify the driver that he can enter L2 / L3 autonomous driving by long pressing the autonomous driving button. L2 / L3 both need to check geo-fences, so it is necessary to first determine whether the vehicle's current location is within the preset geo-fence. If so, monitor whether the preset autonomous driving mode requirements are met. Geo-fencing is a new application of LBS, which is to use a virtual fence to enclose a virtual geographic boundary. When the vehicle enters, leaves, or moves in a specific geographic area, it can receive automatic notifications and warnings.

[0023] like Figure 4 As shown in the figure, rainfall detection in an ODD weather environment is used as an example to describe the rainfall detection method and strategy. The rainfall detection method primarily uses the vehicle's intelligent camera to collect rain light. Based on a visual rain segmentation algorithm and a deep learning rain measurement algorithm, the rainfall amount is calculated. Overall, the method implements intelligent, visual classification of rainfall in the scene, which includes five categories: sunny, light rain, moderate rain, heavy rain, and torrential rain. The method uses video data captured by the intelligent vehicle's camera as input, recording rainfall data at different locations and time domains. This video data is then fed into the "rainfall classification algorithm module," which involves object extraction, segmentation, clustering, and recognition in the environment. First, a rainfall segmentation algorithm based on superpixel neighborhood alignment is implemented: the selected algorithm measures the regional intensity changes of pixel values across different time domains in the rainfall video to obtain rain segmentation results for each frame. Through iterative optimization training, the recognition accuracy is continuously improved, ultimately achieving the desired recognition results. The algorithm implementation consists of the following seven steps.

[0024] Step 11, collecting rainfall video and intercepting a fixed number of frames as input data; Step 12: Pixels are grouped according to their similarity in pixel features. The color image is converted into a 5-dimensional feature vector through linear iterative clustering. A distance metric is constructed for the vector, and pixels are locally clustered to generate a superpixel block map. A superpixel block map is generated through intra-frame superpixel block recognition. That is, pixels are grouped using the feature similarity between pixels (such as similar texture, color brightness, etc.). Through simple linear iterative clustering, the color image is converted into a 5-dimensional feature vector in CIELAB color space and XY coordinates. Then, a distance metric is constructed for the 5-dimensional feature vector, and the image pixels are locally clustered. Ultimately, compact and approximately uniform superpixels are generated, which has high performance in terms of computing speed, object contour preservation, and superpixel shape. A small number of superpixels replaces a large number of pixels to express image features.

[0025] Step 13: Select the middle frame as the key frame and use it as the reference position frame, and form a group of partitioned regions with the superpixel block images of the two frames before and after it; Step 14: construct a loss function to measure the difference between each superpixel block in the superpixel block map and its neighboring blocks in the two previous and next frames, and use the neighboring block with the smallest difference as the candidate alignment area; Step 15: Compare the pixel differences between the superpixel block and the candidate alignment region. If the difference between a pixel in the superpixel block and a pixel corresponding to the candidate alignment region is greater than a threshold, the pixel is determined to be a rain pixel. This judgment is repeated for each pixel in the video to obtain a rain segmentation result. In step 16, the rainwater segmentation results are processed through a deep learning neural network to obtain a rainfall classification result, thereby completing rainfall detection.

[0026] Step 2: If the requirements for L2 autonomous driving mode are met, the chassis, body, and cabin components are monitored for changes in state to determine whether there are any faults. If there are no faults, the requirements for L3 autonomous driving mode are met, and the process goes to Step 3. If there are faults, the fault condition is fed back to the vehicle's intelligent driving system, and the process goes to Step 4. Step 2 also includes: If the L2 autonomous driving mode requirements are not met, the autonomous driving mode cannot be activated. If the driver activates the autonomous driving mode when neither the L2 nor L3 activation conditions are met, neither L3 nor L2 shall be activated. The system shall prompt the driver to try again after displaying the steering wheel icon on the human-machine interface.

[0027] Step 3: Send an L3 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to L3 autonomous driving mode, and then go to step 7; When the system detects that all conditions for enabling the L3 autonomous driving mode are met, the system should notify the driver that the L3 autonomous driving mode can be used through the human-computer interaction interface. The L3 autonomous driving mode startup conditions and L3 level ODD are shown in Table 1.

[0028] Table 1 L3 ODD:

[0029] Step 4: The intelligent driving system determines the fault condition. If it is a serious fault, go to step 5; if it is a minor fault, go to step 6. Step 5: Send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, switch to L3 autonomous driving mode and execute the preset minimum risk strategy; Step 6: Send an L2 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to the L2 autonomous driving mode, and then go to step 7; When the system determines that the conditions exceed the L3 ODD but all the conditions for enabling the L2 autonomous driving mode are met, the system should notify the driver through the human-computer interaction interface that the L2 autonomous driving mode is available. The L2 autonomous driving mode activation conditions and L2 ODD are shown in Table 2.

[0030] Table 2 L2 ODD:

[0031] Step 7: After issuing an L2 / L3 autonomous driving request, if a start signal issued by the driver is detected, the corresponding autonomous driving mode is started.

[0032] Step 8: When the vehicle receives an exit signal from the driver, it exits the automatic driving system.

[0033] The driver's exit signal is specifically issued when the driver presses a designated button, when the driver's intervention in the vehicle's steering exceeds a preset threshold, or when the driver presses the brake pedal. The start / exit module is also configured to exit the automated driving system upon receipt of the driver's exit signal. Typically, both high- and low-level intelligent driving systems share the same start / exit button.

[0034] This invention provides a method for switching intelligent driving modes based on autonomous driving. It primarily explains how to activate and exit an L3 autonomous driving system and switch between high- and low-level intelligent driving modes. This method ensures that L3 vehicles can smoothly activate and exit intelligent driving functions, while also enabling switching between high- and low-level intelligent driving functions, thereby enhancing the user's autonomous driving experience. This method addresses the existing issues of failing to monitor the driver's ability to take over the vehicle, failing to control the exit of L3 autonomous driving, and failing to consider switching between high- and low-level intelligent driving modes.

[0035] Example 3: The present invention provides a method for switching intelligent driving modes based on autonomous driving, which is applied to L3 autonomous driving vehicles. Figure 1 、 3 As shown, the method includes the following steps.

[0036] Step 1: After the vehicle is powered on, it continuously checks whether the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status meet the preset L2 autonomous driving mode requirements; Step 2: If the requirements for L2 autonomous driving mode are met, the chassis, body, and cabin components are monitored for changes in state to determine whether there are any faults. If there are no faults, the requirements for L3 autonomous driving mode are met, and the process goes to Step 3. If there are faults, the fault condition is fed back to the vehicle's intelligent driving system, and the process goes to Step 4. Step 3: Send an L3 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to L3 autonomous driving mode, and then go to step 7; Step 4: The intelligent driving system determines the fault condition. If it is a serious fault, go to step 5; if it is a minor fault, go to step 6. Step 5: Send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, switch to L3 autonomous driving mode and execute the preset minimum risk strategy; Step 6: Send an L2 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to the L2 autonomous driving mode, and then go to step 7; Step 7: After issuing an L2 / L3 autonomous driving request, if a start signal issued by the driver is detected, the corresponding autonomous driving mode is started.

[0037] Step 7 further includes: if no start signal is detected, continuously sending the L2 / L3 autonomous driving request until the driver responds to the L2 / L3 autonomous driving request, or until the request continues for a preset duration, then ceasing to send the L2 autonomous driving request. Detecting the driver's response to the L2 / L3 autonomous driving request includes detecting the driver issuing a start signal or exit signal. In other words, whenever the system detects that the L2 / L3 autonomous driving mode is available, it should send an autonomous driving mode request, notifying the driver via the human-machine interface that the L2 / L3 autonomous driving mode is available. If the system detects the driver long-pressing the autonomous driving button during the L2 / L3 autonomous driving mode request, it should enable the corresponding autonomous driving mode. There are two types of autonomous driving mode requests, depending on the duration of the request: a continuous autonomous driving mode request that persists until the driver responds, and an autonomous driving mode request with a fixed duration. Autonomous driving mode requests with a fixed duration are only used when switching from L3 to L2 autonomous driving mode.

[0038] This invention provides a method for switching intelligent driving modes based on autonomous driving. It primarily explains how to activate and exit an L3 autonomous driving system and switch between high- and low-level intelligent driving modes. This method ensures that L3 vehicles can smoothly activate and exit intelligent driving functions, while also enabling switching between high- and low-level intelligent driving functions, thereby enhancing the user's autonomous driving experience. This method addresses the existing issues of failing to monitor the driver's ability to take over the vehicle, failing to control the exit of L3 autonomous driving, and failing to consider switching between high- and low-level intelligent driving modes.

[0039] In summary, the embodiment of the present invention provides a method and system for switching intelligent driving modes based on automatic driving. After completing the vehicle power-on self-test, the intelligent driving system will monitor in real time whether the L2 / L3 startup conditions meet the requirements; when the system determines that the L2 / L3 startup conditions are met, it will send an automatic driving mode request; when the driver presses the automatic driving button for a long time, the system combines the real-time monitoring status of the L2 / L3 startup conditions and different intelligent driving activation scenarios to enter the corresponding intelligent driving mode; when appropriate conditions appear and the driver responds correctly to the system request, switching between L3 and L2 driving modes is allowed; the driver can independently choose to exit the intelligent driving mode. This technical solution can switch between high and low-level intelligent driving modes for different L2 / L3 driving mode switching scenarios, further improving the safety and user experience of automatic driving, and promoting the rapid development of automatic driving technology.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for switching intelligent driving modes based on autonomous driving, applied to L3 autonomous driving vehicles, characterized in that: The method comprises: Step 1: After the vehicle is powered on, it continuously checks whether the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status meet the preset L2 autonomous driving mode requirements; Step 2: If the requirements for L2 autonomous driving mode are met, the chassis, body, and cabin components are monitored for changes in state to determine whether there are any faults. If there are no faults, the requirements for L3 autonomous driving mode are met, and the process goes to Step 3. If there are faults, the fault condition is fed back to the vehicle's intelligent driving system, and the process goes to Step 4. Step 3: Send an L3 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to L3 autonomous driving mode, and then go to step 7; Step 4: The intelligent driving system determines the fault condition. If it is a serious fault, go to step 5; if it is a minor fault, go to step 6. Step 5: Send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, switch to L3 autonomous driving mode and execute the preset minimum risk strategy; Step 6: Send an L2 autonomous driving request to the human-computer interaction interface to prompt the driver to switch to the L2 autonomous driving mode, and then go to step 7; Step 7: After issuing an L2 / L3 autonomous driving request, if a start signal issued by the driver is detected, the corresponding autonomous driving mode is started.

2. The method for switching intelligent driving modes based on automatic driving according to claim 1, characterized in that: The method further includes: step 8, exiting the automatic driving system when the vehicle receives an exit signal from the driver.

3. The method for switching intelligent driving modes based on automatic driving according to claim 2, characterized in that: Specifically, the driver sends the exit signal when the driver touches a designated button, or when the driver's intervention in the vehicle steering control exceeds a preset threshold, or when the driver steps on the brake pedal.

4. The method for switching intelligent driving modes based on automatic driving according to claim 1, characterized in that: The step 1 also includes: determining whether the current position of the vehicle is within a preset geographic fence, and if so, monitoring whether the preset automatic driving mode requirements are met.

5. The method for switching intelligent driving modes based on automatic driving according to claim 1, characterized in that: Step 2 also includes: if the L2 autonomous driving mode requirements are not met, the autonomous driving mode cannot be started.

6. The method for switching intelligent driving modes based on automatic driving according to claim 1, characterized in that: Step 7 also includes: if no start signal is detected, continuing to send the L2 / L3 autonomous driving request until the driver is detected to respond to the L2 / L3 autonomous driving request, or stopping sending the L2 autonomous driving request after the request continues for a preset time period.

7. The method for switching intelligent driving modes based on automatic driving according to claim 6, characterized in that: The detecting that the driver responds to the L2 / L3 autonomous driving request includes: detecting that the driver sends a start signal or an exit signal.

8. The method for switching intelligent driving modes based on automatic driving according to claim 1, characterized in that: The weather environment detection includes rainfall detection, and the rainfall detection method includes: Step 11, collecting rainfall video and intercepting a fixed number of frames as input data; Step 12: Pixels are grouped according to their similarity in pixel features. The color image is converted into a 5-dimensional feature vector through linear iterative clustering. A distance metric is constructed for the vector, and pixels are locally clustered to generate a superpixel block map. Step 13: Select the middle frame as the key frame and use it as the reference position frame, and form a group of partitioned regions with the superpixel block images of the two frames before and after it; Step 14: construct a loss function to measure the difference between each superpixel block in the superpixel block map and its neighboring blocks in the two previous and next frames, and use the neighboring block with the smallest difference as the candidate alignment area; Step 15: Compare the pixel differences between the superpixel block and the candidate alignment region. If the difference between a pixel in the superpixel block and a pixel corresponding to the candidate alignment region is greater than a threshold, the pixel is determined to be a rain pixel. This judgment is repeated for each pixel in the video to obtain a rain segmentation result. In step 16, the rainwater segmentation results are processed through a deep learning neural network to obtain a rainfall classification result, thereby completing rainfall detection.

9. A system for implementing the method for switching intelligent driving modes based on autonomous driving according to claims 1-8, characterized in that: The system comprises: The detection module is used to continuously monitor the vehicle status, road conditions, weather conditions, road facilities, target conditions, and driver status after the vehicle is powered on to determine whether they meet the preset L2 autonomous driving mode requirements. If so, the module determines whether there are any faults in the chassis, body, and cabin based on changes in the status of the opponent components, and reports the faults to the vehicle's intelligent driving system. The intelligent driving system is connected to the detection module. If there is no fault, it sends an L3 autonomous driving request to the human-computer interaction interface. If there is a fault, it will judge and distinguish the fault situation. If it is judged to be a serious fault, it will send a manual takeover request to the human-computer interaction interface. If no one responds to the takeover request, it will switch to L3 autonomous driving mode and execute the preset minimum risk strategy. If it is judged to be a minor fault, it will send an L2 autonomous driving request to the human-computer interaction interface. The start / exit module is connected to the intelligent driving system and is used to activate the corresponding autonomous driving mode if a start signal from the driver is detected after an L2 / L3 autonomous driving request is issued; The human-computer interaction interface is connected to the intelligent driving system and is used to display L2 / L3 autonomous driving requests or manual takeover requests to prompt the driver to switch to L2 / L3 autonomous driving mode or manual driving mode.

10. The intelligent driving mode switching system based on automatic driving according to claim 9, characterized in that: The start / exit module is also used to exit the automatic driving system when the vehicle receives an exit signal from the driver.

Images