Autonomous driving voice flexible intervention control method, system, device and medium

Abstract

The present disclosure provides an automatic driving voice flexible intervention control method, system, device and medium. The method comprises: in the process of automatic driving of a vehicle, obtaining driving environment data under a current working condition and an intervention driving voice instruction issued by a user; performing hierarchical verification on the intervention driving voice instruction to filter external noise and invalid instructions, and obtaining a target voice instruction; checking whether the vehicle meets a preset safe passing condition under the current working condition according to the current driving environment data; and after checking that the safe passing condition is met, controlling the vehicle to automatically drive according to the target voice instruction and the current working condition. Through the multi-level voice verification mechanism and the safe passing condition review, the present disclosure realizes precise, safe and flexible intervention on the automatic driving vehicle, effectively solves the unnecessary parking problem caused by environmental perception misidentification, and meets the individual driving needs of the user.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving control technology, and more specifically, to autonomous driving voice flexible intervention control methods, systems, devices, and media. Background Technology

[0002] With the rapid development of autonomous driving technology, driverless vehicles have been gradually promoted and applied in low-speed scenarios such as scenic spot sightseeing, park shuttles, and community logistics. At present, the autonomous driving function of mainstream driverless vehicles mainly relies on environmental perception modules such as cameras, lidar, and millimeter-wave radar to collect surrounding road condition information. The autonomous driving decision module combines the preset path planning algorithm to output driving commands, and controls the vehicle's power system, steering system, and braking system through the vehicle controller (VCU) or a similar central control unit to achieve basic operations such as constant speed driving, obstacle avoidance and parking, and path tracking.

[0003] In scenarios such as low-speed sightseeing and park shuttle services, existing technologies mainly have the following three problems: First, the environmental perception module has a misidentification problem, which causes the vehicle to stop unnecessarily when it encounters certain objects, affecting the continuity of driving.

[0004] Secondly, the vehicle's speed adjustment, detour and other operations strictly rely on preset route parameters and decision-making logic, making it impossible for users to intervene in the vehicle's behavior in real time according to actual needs, thus failing to meet personalized driving requirements.

[0005] Third, although some driverless vehicles have integrated voice control functions, there is a problem of voice commands being triggered erroneously in scenarios with multiple passengers and complex environmental noise.

[0006] In summary, existing technologies are insufficient to meet the multiple requirements of low-speed sightseeing scenarios, such as driving continuity, user interaction flexibility, and voice control reliability, while ensuring the safety of autonomous driving. Summary of the Invention

[0007] The technical problem to be solved by this disclosure is to provide a method, system, device and medium for flexible voice intervention control of autonomous driving, which aims to solve at least one of the above-mentioned technical problems.

[0008] In a first aspect, the technical solution disclosed herein to solve the above-mentioned technical problems is as follows: an autonomous driving voice-based flexible intervention control method, applied to a vehicle including a vehicle controller, the method comprising: During the autonomous driving process, data on the current driving environment under the current operating conditions and the user's voice commands for intervention are acquired. The intervention driving voice command is verified in a hierarchical manner to filter out external noise and invalid commands in the intervention driving voice command and obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. Determine whether the vehicle meets the preset safe passage conditions under the current operating conditions based on the current driving environment data; After determining that the vehicle meets the safe passage conditions, the system controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions.

[0009] The beneficial effects of this disclosure are as follows: First, through a multi-level voice command verification mechanism, external noise interference and invalid commands from unauthorized users in complex environments such as scenic spots and parks are effectively filtered out, significantly improving the accuracy and security of voice control and avoiding driving safety hazards caused by noise interference or accidental command triggering. Second, after successful verification, a safe passage condition judgment based on current driving environment data is introduced to ensure that the execution of user voice commands is based on the premise of physical environment safety, forming a dual guarantee mechanism of "user intent and environmental safety" to avoid safety risks caused by user misjudgment or system misidentification. Third, by combining the execution of user commands with the current operating conditions, real-time flexible intervention in operations such as vehicle speed adjustment and detour direction is realized, meeting the personalized needs of users in the process of autonomous driving of the vehicle and improving the user experience.

[0010] Based on the above technical solution, the present disclosure can be further improved as follows.

[0011] Furthermore, the above-mentioned hierarchical verification of the intervention driving voice commands filters out external noise and invalid commands from the intervention driving voice commands to obtain the target voice commands, including: Determine whether the signal-to-noise ratio (SNR) of the intervention driving voice command is not less than the preset SNR threshold. If the SNR is not less than the SNR threshold, determine that the intervention driving voice command has passed the noise reduction verification. The voice command that has passed the noise reduction verification is matched with the command words in the preset keyword library. The matching degree of the obtained keyword is determined to be no less than the preset matching degree threshold. If the matching degree of the keyword is no less than the matching degree threshold, the voice command that has passed the noise reduction verification is determined to have passed the keyword matching verification. Based on the current working conditions, determine whether the command word corresponding to the voice command verified by keyword matching matches the current working conditions. If the command word matches the current working conditions, determine that the voice command verified by keyword matching has passed the working condition matching verification. The voice commands that have passed the working condition matching verification are matched with the preset voice features of the main and secondary personnel for permission verification. It is determined whether the obtained permission matching degree is not less than the preset permission threshold. If the permission matching degree is not less than the permission threshold, the voice commands that have passed the working condition matching verification are deemed to have passed the permission verification. The voice command that has passed the permission verification is identified as the target voice command.

[0012] Furthermore, before obtaining the intervention driving voice command, the method also includes: In response to the current driving environment data meeting the preset voice intervention trigger conditions, a voice intervention prompt is issued. The voice intervention trigger conditions include: the autonomous driving decision module outputs a parking decision, or outputs a detour request, or receives a user's active wake-up command.

[0013] Furthermore, the aforementioned current driving environment data includes at least one environmental parameter associated with the target voice command. Determining whether the vehicle meets preset safe passage conditions under the current operating conditions based on the current driving environment data includes: Based on the target voice command, obtain the preset security verification rules corresponding to the command type of the target voice command; According to the safety verification rules, at least one environmental parameter in the current driving environment data is verified in conjunction with the current operating condition of the vehicle. If the verification passes, the vehicle is determined to meet the safe passage conditions under the current operating conditions; if the verification fails, the vehicle is determined to not meet the safe passage conditions under the current operating conditions.

[0014] Furthermore, according to the safety verification rules, at least one environmental parameter in the current driving environment data is verified in conjunction with the vehicle's current operating condition, including: If the current working condition is a waiting-to-intervene working condition where a parking decision is triggered by a suspected obstacle, and the target voice command is a command to continue driving, the basic parameters of the obstacle in the current driving environment data are obtained, and the road conditions corresponding to the basic parameters of the obstacle are verified to meet the safe passage conditions. If the current driving condition is constant speed driving condition and the target voice command is a deceleration command, obtain the current vehicle speed data in the current driving environment data and verify whether the target vehicle speed after deceleration is within the preset safe vehicle speed range. If the current operating condition is a waiting situation for oncoming traffic or a situation where the passage space is limited, and the target voice command is a detour command, obtain the lateral space data corresponding to the detour direction from the current driving environment data, and verify whether the lateral space data meets the vehicle's safe passage distance requirements.

[0015] Furthermore, after determining that the vehicle meets the safe passage conditions, the system controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions, including: Based on the current operating conditions and the target voice command, obtain the target control strategy corresponding to the target voice command; After determining that the vehicle meets the conditions for safe passage, the system controls the vehicle to perform the corresponding autonomous driving operation according to the target control strategy.

[0016] Furthermore, the above-mentioned acquisition of the target control strategy corresponding to the target voice command based on the current operating conditions and the target voice command includes: When the current working condition is a parking decision triggered by a suspected obstacle and the target voice command is a continue driving command, the target control strategy obtained is: cancel the original parking decision and control the vehicle to maintain the preset normal cruising speed and continue driving. When the current operating condition is constant speed driving and the target voice command is a deceleration command, the target control strategy obtained is: control the vehicle to reduce the current speed to the preset low speed range. When the current operating condition is parking and the target voice command is to resume driving, the target control strategy is to control the vehicle to re-enter the original driving route and restore it to the preset normal cruising speed. When the current working condition is waiting for oncoming traffic or limited passage space, and the target voice command is a detour command, the target control strategy is: control the vehicle to complete the detour operation in the direction corresponding to the detour command, and automatically return to the original driving route after detour and restore the preset normal cruise speed. When the current operating condition is a non-parking driving condition, and the target voice command is a parking command, the obtained target control strategy is: control the vehicle's braking system to activate the parking control unit to perform smooth parking.

[0017] Furthermore, the method also includes: During the process of controlling the vehicle to perform corresponding autonomous driving operations according to the target control strategy, feedback information corresponding to the target voice command and / or the target control strategy is generated. Feedback information will be provided to users in real time through voice prompts and / or interface displays.

[0018] Secondly, in order to solve the above-mentioned technical problems, this disclosure also provides an autonomous driving voice flexible intervention control system, applied to a vehicle including a vehicle controller, the system comprising: The acquisition module is used to acquire current driving environment data and user-issued voice commands for intervention during the autonomous driving process. The verification module is used to perform hierarchical verification of the intervention driving voice command, filter out external noise and invalid commands in the intervention driving voice command, and obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. The automatic decision-making module is used to determine whether the vehicle meets the preset safe passage conditions under the current operating conditions based on the current driving environment data. The control module is used to control the vehicle to perform autonomous driving based on the target voice command and the current operating conditions after determining that the vehicle meets the safe passage conditions.

[0019] Thirdly, in order to solve the above-mentioned technical problems, this disclosure also provides a vehicle that includes the autonomous driving voice flexible intervention control system described in the second aspect.

[0020] Fourthly, in order to solve the above-mentioned technical problems, this disclosure also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the autonomous driving voice flexible intervention control method of this application.

[0021] Fifthly, in order to solve the above-mentioned technical problems, this disclosure also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the autonomous driving voice flexible intervention control method of this application.

[0022] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments of this disclosure will be briefly introduced below.

[0024] Figure 1 A flowchart illustrating an embodiment of the autonomous driving voice-based flexible intervention control method provided in this disclosure; Figure 2 This is a schematic diagram of the structure of an autonomous driving voice flexible intervention control system provided in one embodiment of the present disclosure; Figure 3 A flowchart illustrating another method for voice-based flexible intervention control of autonomous driving, provided as an embodiment of this disclosure; Figure 4 This is a schematic diagram of the structure of an autonomous driving voice flexible intervention control system provided in one embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present disclosure. Detailed Implementation

[0025] The principles and features of this disclosure are described below. The examples given are for illustrative purposes only and are not intended to limit the scope of this disclosure.

[0026] The technical solutions of this disclosure and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this disclosure will now be described with reference to the accompanying drawings.

[0027] The acquisition process of the data involved in this disclosure follows the principles of legality, legitimacy, and necessity. Based on the explicit authorization and consent of the users, only the minimum necessary information required to achieve the purpose is collected, and data security protection obligations are fulfilled in accordance with the law.

[0028] The solutions provided in this disclosure can be applied to any scenario requiring autonomous driving voice interaction control, such as low-speed sightseeing or park shuttles, and are particularly suitable for application scenarios with complex environments and high noise levels, such as scenic spots and parks.

[0029] This disclosure provides a possible implementation, such as... Figure 1 The diagram shows a flowchart of a method for flexible voice intervention control in autonomous driving. This method can be executed by any electronic device, such as a vehicle control unit (VCU). For ease of description, the method provided in this disclosure will be described below using the vehicle control unit (VCU) as an example. Figure 1 The flowchart shown indicates that the method may include the following steps: S10, during the autonomous driving process, acquires current driving environment data under the current working conditions and the user's intervention driving voice commands; S20, perform hierarchical verification of the intervention driving voice command, filter out external noise and invalid commands in the intervention driving voice command, and obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. S30 determines whether the vehicle meets the preset safe passage conditions under the current operating conditions based on the current driving environment data; After determining that the vehicle meets the safe passage conditions, S40 controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions.

[0030] The method disclosed herein, firstly, effectively filters out external noise interference and invalid commands from unauthorized users in complex environments such as scenic spots and parks through a multi-level voice command verification mechanism, significantly improving the accuracy and security of voice control and avoiding driving safety hazards caused by noise interference or mis-triggered commands. Secondly, after successful verification, a safe passage condition judgment based on current driving environment data is introduced to ensure that the execution of user voice commands is premised on physical environmental safety, forming a dual guarantee mechanism of "user intent and environmental safety" to avoid safety risks caused by user misjudgment or system misidentification. Thirdly, by combining the execution of user commands with the current operating conditions, real-time flexible intervention in operations such as vehicle speed adjustment and detour direction is achieved, meeting the personalized needs of users in the process of autonomous driving and improving the user experience.

[0031] The present disclosure will be further described below with reference to the following specific embodiments. In this embodiment, the autonomous driving voice flexible intervention control method provided by the present solution can be applied to... Figure 2 The diagram shown below illustrates the structure of a voice-activated flexible intervention control system for autonomous driving. (The following is a description of the system in conjunction with...) Figure 2 The system structure diagram shown further illustrates the autonomous driving voice-based flexible intervention control method provided in this embodiment. The method may include the following steps: S10, during the autonomous driving process, acquires current driving environment data under the current working conditions and the user's intervention driving voice commands; Specifically, this step marks the beginning of the voice intervention control process initiated by the Vehicle Control Unit (VCU). When the vehicle is driving in autonomous driving mode in low-speed sightseeing scenarios such as scenic spots and parks, the VCU continuously acquires current driving environment data from environmental perception modules (including cameras, LiDAR, millimeter-wave radar, etc.), and simultaneously monitors the voice signals in the passenger area of ​​the vehicle in real time through the voice recognition module. Specifically, the voice signals in the passenger area of ​​the vehicle can be collected through the microphone array built into the voice recognition module.

[0032] The vehicle can be a sightseeing vehicle, such as a small, unmanned sightseeing vehicle with four seats. The current operating condition refers to the vehicle's current operating status, such as a waiting-to-stop condition triggered by a suspected obstacle, a constant-speed driving condition, a waiting-for-on-coming-car condition, or a condition with limited space. Current driving environment data refers to various environmental information related to vehicle driving safety collected by the environmental perception module, including but not limited to basic parameters of obstacles, current vehicle speed data, and lateral space data. The user can be a passenger in the vehicle, referring to the main passenger or driver with operational authority within the vehicle, typically a primary passenger or driver with pre-recorded voice characteristics. Intervention voice commands refer to voice commands issued by passengers in the vehicle based on actual observation and needs, intended to intervene in the vehicle's autonomous driving behavior, such as "continue driving," "decelerate," or "turn right." These can be obtained through a voice recognition module. Specifically, the input end of the voice recognition module can be connected to the in-vehicle microphone array to receive intervention voice commands issued by the user, and the output end is connected to the vehicle control unit (VCU). The voice recognition module executes step S20 and then sends the target voice command, obtained after hierarchical verification, to the VCU for subsequent processing.

[0033] Optionally, the aforementioned environmental perception module can be composed of multiple sensors, specifically including: Cameras are used to collect visual image information in front of a vehicle and identify the appearance features of lane lines, traffic signs, pedestrians, and various obstacles. LiDAR (Light Detection and Ranging) generates three-dimensional point cloud data of the environment around a vehicle by emitting laser beams and receiving reflected signals, accurately obtaining basic parameters such as the position, size, and outline of obstacles. Millimeter-wave radar uses electromagnetic waves in the millimeter-wave band to detect targets and obtain the relative distance, relative speed and azimuth angle of obstacles in real time. It can maintain stable detection performance, especially in adverse weather conditions such as rain, fog and dust.

[0034] The aforementioned multiple sensors work together to send the collected raw data to the data fusion unit of the environmental perception module. After fusion processing, structured current driving environment data is generated, including basic parameters of obstacles, current vehicle speed data, and lateral space data, and then transmitted to the VCU for subsequent decision-making. Through this multi-sensor fusion configuration, the system can comprehensively and accurately perceive the vehicle's surrounding environment, providing reliable data support for subsequent voice command verification and safe passage condition checks.

[0035] Optionally, before obtaining the intervention driving voice command, the method further includes: In response to the current driving environment data meeting the preset voice intervention trigger conditions, a voice intervention prompt is issued. The voice intervention trigger conditions include: the autonomous driving decision module outputs a parking decision, or outputs a detour request, or receives a user's active wake-up command.

[0036] Voice intervention prompts refer to the voice information issued by the vehicle through the voice synthesis module to passengers when any of the above conditions are triggered during autonomous driving. This information informs the user of the current vehicle status and prompts them to intervene via voice. For example, when the environmental perception module misidentifies a floating plastic bag as an obstacle and makes a parking decision, it can issue a voice prompt saying "Obstacle detected, parking is imminent"; when it issues a detour request due to oncoming traffic or limited space, it can issue a voice prompt saying "Passage space ahead is limited, please confirm the detour direction".

[0037] After receiving the aforementioned voice intervention prompts, users can, based on their own observation and judgment of the actual road conditions (such as confirming that the height of the plastic bag is lower than the chassis and that there is sufficient side space), proactively issue intervention driving voice commands that match the current operating conditions, such as "continue driving" or "detour to the right," thereby achieving real-time flexible intervention in vehicle driving behavior.

[0038] S20, perform hierarchical verification of the intervention driving voice command, filter out external noise and invalid commands in the intervention driving voice command, and obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. After receiving the user's voice command to intervene in driving, due to the complex environmental noise in open settings such as scenic spots and parks (such as wind noise and tourists' chatter), and the presence of multiple passengers in the vehicle, directly executing the voice command to intervene in driving may pose a risk of accidental triggering or unauthorized operation. Therefore, this step employs a multi-level verification mechanism to ensure that only genuine, valid commands that are appropriate for the current scenario are adopted.

[0039] It should be noted that the aforementioned hierarchical verification mechanism is designed to address the objective characteristics of low-speed sightseeing scenarios, such as multiple passengers and complex environmental noise. Through a four-layer progressive verification process—noise reduction verification, keyword matching verification, operating condition matching verification, and authorization verification—it effectively filters out background noise interference such as wind noise and passenger chatter, preventing accidental triggering of voice commands due to noise, and preventing unauthorized personnel from misoperating the vehicle, thus ensuring driving safety.

[0040] In a specific embodiment, in S20 above, the intervention driving voice command is subjected to hierarchical verification to filter out external noise and invalid commands in the intervention driving voice command, thereby obtaining the target voice command, including: S201, determine whether the signal-to-noise ratio of the intervention driving voice command is not less than the preset signal-to-noise ratio threshold. If the signal-to-noise ratio is not less than the signal-to-noise ratio threshold, determine that the intervention driving voice command has passed the noise reduction verification. S202, the voice command that has passed the noise reduction verification is matched with the command words in the preset keyword library, and it is determined whether the obtained keyword matching degree is not less than the preset matching degree threshold. If the keyword matching degree is not less than the matching degree threshold, the voice command that has passed the noise reduction verification is determined to have passed the keyword matching verification. S203, based on the current operating conditions, determine whether the instruction word corresponding to the voice command verified by keyword matching matches the current operating conditions. If the instruction word matches the current operating conditions, determine that the voice command verified by keyword matching has passed the operating condition matching verification. Here, determining whether the instruction word matches the current operating conditions means that after the intervention driving voice command has passed the noise reduction verification and keyword matching verification, further verify whether the driving intention expressed by the instruction word is consistent with the current driving state of the vehicle (i.e., the current operating conditions) and has the logical rationality for execution.

[0041] S204, the voice command that has passed the working condition matching verification is matched with the preset voice features of the main and passenger personnel for permission, and it is determined whether the obtained permission matching degree is not less than the preset permission threshold. If the permission matching degree is not less than the permission threshold, the voice command that has passed the working condition matching verification is determined to have passed the permission verification. S205, the voice command that has passed the authorization verification is identified as the target voice command.

[0042] The keyword library is a pre-stored collection of all legal voice command words, used to match and recognize voice commands that have passed noise reduction verification during the voice command verification process. This keyword library pre-sets corresponding command words based on various driving scenarios the vehicle may encounter (such as obstacle misidentification, speed adjustment, and detouring to avoid oncoming traffic), such as "continue driving," "decelerate," and "detour to the right." This ensures that only voice commands with a matching degree of at least a preset threshold with the command words in the library can pass the keyword matching verification, thereby effectively filtering out invalid commands and irrelevant voice interference.

[0043] Keyword matching score refers to the quantified value of similarity or matching degree obtained when comparing a noise-reduced voice command with command words in a preset keyword library during the keyword matching verification stage of voice command verification. This matching score is usually calculated by a speech recognition algorithm and expressed as a percentage or a value between 0 and 1, used to measure the degree of conformity between the user's voice and the standard command words. When the keyword matching score reaches or exceeds a preset matching score threshold (such as 90%), the voice command is deemed to have passed the keyword matching verification, indicating that the user's intention command has been successfully recognized.

[0044] The voice characteristics of the primary occupant refer to the unique voiceprint biometric data of the main occupant authorized to operate the vehicle, which is collected and stored in advance. This characteristic data is generated by analyzing unique parameters such as pitch, timbre, speech rate, and pronunciation habits in the occupant's voice signal. It is used to compare the user issuing the voice command with the pre-stored voiceprint model of the primary occupant during the authorization verification process. The authorization matching degree refers to the similarity metric obtained by comparing the voice command that has passed the working condition matching verification with the preset voice characteristics of the primary occupant during the authorization verification process. This matching degree is usually calculated by a voiceprint recognition algorithm and expressed as a percentage or a value between 0 and 1. It is used to measure the degree to which the issuer of the current voice command matches the identity of the authorized primary occupant. When the authorization matching degree reaches or exceeds a preset authorization threshold (such as 85%), the voice command is determined to have passed authorization verification, confirming that the current command comes from the primary occupant with operating authority, thereby effectively preventing unauthorized personnel from operating the vehicle and ensuring driving safety.

[0045] Furthermore, prior to S201, the method also includes: S200 filters out external environmental noise in the intervention driving voice command, generates a pre-processed voice signal, and then replaces the pre-processed voice signal with the intervention driving voice command in S201 to perform subsequent processing.

[0046] As an example, the above hierarchical verification process includes: First, performing noise reduction verification to determine if the signal-to-noise ratio (SNR) of the preprocessed speech signal is not less than a preset SNR threshold (e.g., 35dB). If it is not less, the noise reduction verification is passed; otherwise, it is failed. Second, performing keyword matching verification to match the speech signal that passed noise reduction verification with command words (such as "continue driving," "decelerate," "detour to the right," etc.) in a preset keyword library. If the keyword matching degree is not less than a preset matching degree threshold (e.g., 90%), the keyword matching verification is passed; otherwise, it is failed. Third, performing working condition matching verification to determine if the command word corresponding to the speech signal that passed keyword matching verification matches the current working condition. If the command word matches the current working condition, the working condition matching verification is passed; otherwise, it is failed. For example, in a waiting-to-stop condition triggered by a suspected obstacle, the "continue driving" command is matched, while in a constant-speed driving condition, the "decelerate" command is matched. Finally, permission verification is performed by matching the voice signal that has passed the working condition matching verification with the preset voice characteristics of the driver and passenger. If the permission matching degree is not less than the preset permission threshold (e.g., 85%), the permission verification is deemed successful. The voice command that passes all four layers of verification is finally identified as the target voice command.

[0047] S30 determines whether the vehicle meets the preset safe passage conditions under the current operating conditions based on the current driving environment data; After receiving a verified target voice command, the VCU does not execute the command immediately. Instead, it performs a safety check based on current driving environment data. This is to ensure that the execution of user commands is premised on physical environment safety, forming a dual protection mechanism of "user intent + environmental safety".

[0048] It should be noted that the safe passage condition verification mechanism is designed to address potential misidentification. When the vehicle misidentifies certain objects as obstacles, leading to unnecessary stopping decisions, the user can initiate intervention via voice command. The system will review the safety of the command execution based on current driving environment data. Only if the physical environment confirms that passage is permitted will the user's intervention command be executed, thus ensuring driving safety while meeting the user's personalized driving needs.

[0049] In one specific embodiment, the aforementioned current driving environment data includes at least one environmental parameter associated with the target voice command. In step S30, determining whether the vehicle meets preset safe passage conditions under the current operating conditions based on the current driving environment data includes: S301, Based on the target voice command, obtain the preset security verification rules corresponding to the command type of the target voice command; S302, According to the safety verification rules, at least one environmental parameter in the current driving environment data is verified in combination with the current operating condition of the vehicle. S303: If the verification passes, the vehicle is determined to meet the safe passage conditions under the current operating conditions. Conversely, if the verification fails, the vehicle is determined to not meet the safe passage conditions under the current operating conditions.

[0050] Among them, safety verification rules are a set of pre-defined logical rules for different command types, used to verify whether the execution of the command is safe. For example, for the "continue driving" command, the safety verification rules may include whether the road conditions corresponding to the basic parameters of the obstacle meet the safe passage conditions; for the "decelerate" command, the safety verification rules may include whether the target vehicle speed after deceleration is within the preset safe speed range.

[0051] Optionally, considering the specific current operating conditions and the specific target voice command, one possible implementation of the above S302 includes: S3021, If ​​the current working condition is a working condition that requires intervention due to a suspected obstacle triggering a parking decision, and the target voice command is a command to continue driving, obtain the basic parameters of the obstacle in the current driving environment data, and verify whether the road conditions corresponding to the basic parameters of the obstacle meet the safe passage conditions. S3022, if the current operating condition is constant speed driving condition and the target voice command is deceleration command, obtain the current vehicle speed data in the current driving environment data and verify whether the target vehicle speed after deceleration is within the preset safe vehicle speed range. S3023 If the current operating condition is waiting for oncoming traffic or limited passage space, and the target voice command is a detour command, obtain the lateral space data corresponding to the detour direction in the current driving environment data, and verify whether the lateral space data meets the vehicle's safe passage distance requirements.

[0052] Lateral space data refers to the continuous, unobstructed area on the designated side (left or right) of the detour direction where a vehicle can safely pass, along with its related environmental parameters. Based on these environmental parameters, it is possible to more accurately verify whether the lateral space data meets the safe passing distance requirements for vehicles.

[0053] The intervention condition where a parking decision is triggered by a suspected obstacle refers to a situation where an object that does not actually affect passage (such as a fluttering plastic bag or low shrubs) is mistakenly identified as a real obstacle, causing the vehicle to be about to stop.

[0054] Optionally, for intervention scenarios where a parking decision is triggered by a suspected obstacle, the following situations may exist: In the first scenario, there is only one obstacle, and the obstacle is located in the vehicle's path.

[0055] At this point, the basic parameters of the obstacle include its highest point and its position on the road surface. Verifying whether the road conditions corresponding to the obstacle's basic parameters meet the requirements for safe passage involves the following steps: First, verify whether the lateral distance between the obstacle and the road boundary is greater than the safe passage width of the vehicle. The lateral distance refers to the minimum horizontal distance between the outermost point of the obstacle and the road boundary, perpendicular to the vehicle's direction of travel; the safe passage width is the sum of the vehicle's width and a preset safety margin. If the lateral distance is greater than the safe passage width, it is determined that the vehicle can pass around the obstacle from one side, meeting the safe passage conditions.

[0056] If the lateral distance is not greater than the safe passage width, the detour space is insufficient, and the height parameters of the obstacle need to be further verified. Specifically, it is verified whether the vertical distance between the highest point of the obstacle and the ground is lower than the vehicle's minimum ground clearance. The minimum ground clearance refers to the vertical distance between the lowest point of the vehicle's bottom and the ground. If the obstacle height is lower than the minimum ground clearance, it is determined that the vehicle can pass over the obstacle, meeting the safe passage conditions; otherwise, it is determined that it cannot pass safely, and the vehicle maintains its original stopping decision.

[0057] It should be noted that the lateral distance in the above verification is only valid when the line connecting the obstacle and the road boundary is substantially perpendicular to the vehicle's direction of travel. If the obstacle extends along the direction of travel (such as a long, narrow obstacle (like a fallen tree trunk or a long stone block) extending along the direction of travel instead of lying across the middle of the road), then the detour verification is not applicable, and the vehicle will proceed directly to the height verification.

[0058] In the second scenario, there are two obstacles, located on opposite sides of the vehicle's path.

[0059] At this point, the basic parameters of each obstacle include the highest point of each obstacle, the position of each obstacle on the road surface, and the width data of each obstacle.

[0060] Step 1: Verify whether the vehicle can pass between the two obstacles. Specifically, verify whether the lateral distance between the two obstacles is greater than the safe passage width of the vehicle. If so, the safe passage conditions are met, and the vehicle can pass between the two obstacles. The lateral distance between the two obstacles refers to the horizontal distance between the innermost edges of the two obstacles, perpendicular to the vehicle's direction of travel.

[0061] Step 2: If the vehicle cannot pass between the two obstacles, check whether the vertical distance between the highest point of each obstacle and the ground is lower than the vehicle's minimum ground clearance. If the vertical distance between the highest point of both obstacles and the ground is lower than the vehicle's minimum ground clearance, the vehicle can pass directly over both obstacles simultaneously, meeting the safe passage conditions. If the vertical distance between the highest point of both obstacles and the ground is not lower than the vehicle's minimum ground clearance, the vehicle cannot pass, and the original parking decision is maintained. If only one of the two obstacles has a vertical distance between its highest point and the ground that is lower than the minimum ground clearance, while the other has a vertical distance that is higher than the minimum ground clearance, then proceed to step 3.

[0062] Step 3: First, verify whether it is possible to detour around the high obstacle. Specifically, verify whether the lateral distance between the high obstacle and the road boundary is greater than the safe passage width for vehicles. If so, determine that the vehicle can detour between the high obstacle and the roadside, meeting the safe passage conditions. Here, a high obstacle refers to an obstacle whose highest point is vertically above the ground with a clearance greater than the minimum ground clearance.

[0063] If an alternative route is not possible, the feasibility of crossing the obstacle on one side is verified. Specifically, it is verified whether the width of the low obstacle is less than the width of one tire on the vehicle, and whether the low obstacle is located on the expected travel trajectory of one tire. If these conditions are met, the vehicle is determined to be able to pass over the low obstacle using one tire while the other tire continues to travel normally, thus meeting the safe passage requirements. A low obstacle refers to an obstacle whose highest point is below the minimum ground clearance. The width of one tire refers to the lateral width of the contact area between a single tire and the ground, typically determined by the tire's cross-sectional width, used to verify whether the obstacle is completely beneath the tire. The expected travel trajectory of one tire refers to the projection path of the tire's centerline onto the ground when the vehicle maintains its current direction of travel without active steering. The width of the low obstacle refers to its lateral dimension perpendicular to the vehicle's direction of travel.

[0064] It should be noted that the single-sided obstacle crossing method requires that the low obstacle be positioned exactly on the travel path of one side of the vehicle's tires, and that the vehicle can maintain a straight line to pass through without significant steering. This method cannot be used if the low obstacle is located between the two wheels.

[0065] If neither of the above two methods is feasible, the vehicle is deemed unable to pass safely, and the original parking decision is maintained.

[0066] It should be noted that the "lateral spacing" and "lateral distance from the roadside" mentioned above refer to the horizontal distance perpendicular to the vehicle's direction of travel. The above solutions for handling two obstacles are effective when the two obstacles are arranged side-by-side. If the two obstacles are arranged one in front of the other along the direction of travel (rather than side-by-side), this is not applicable, and each obstacle should be checked individually according to the logic for handling a single obstacle to determine its passability.

[0067] The third scenario involves two or more obstacles located at different points along the driving path.

[0068] At this point, the basic parameters of the obstacles include the highest point of each obstacle, the position of each obstacle on the road surface, the width data of each obstacle, and the relative positional relationship between the obstacles.

[0069] The verification logic is as follows: 1. Traffic zone identification: Specifically, based on the location of obstacles on the road surface, the road ahead of the vehicle is divided into multiple potential traffic zones. Each potential traffic zone must meet the following basic conditions: There are no obstacles within the potential passage area whose height (referring to the vertical distance between the highest point of the obstacle and the ground) is not less than the minimum ground clearance; The width of the potential passage area shall not be less than the safe passage width for vehicles; The potential passage area has a continuous access path in the direction of vehicle travel.

[0070] Potentially passable areas that meet the above basic conditions are identified as candidate passable areas. A potential passable area refers to a continuous space enclosed by obstacle boundaries and road boundaries that a vehicle can potentially traverse. An accessible path refers to a feasible route that allows a vehicle to travel continuously and without obstruction from its current location to its target location in its current direction of travel.

[0071] If a potential passage area exists, proceed with the following steps. If no potential passage area exists, determine that the road conditions corresponding to the basic parameters of the obstacle do not meet the safe passage conditions, and maintain the original parking decision.

[0072] 2. Route feasibility assessment: Specifically, for each identified candidate passage area, its feasibility is assessed, mainly including: 2.1. Area Boundary Verification: Specifically, for each candidate passage area, verify whether the minimum lateral distance between the left and right boundaries of the candidate passage area is greater than the safe passage width for vehicles. If so, the candidate passage area has basic passage space; otherwise, it does not have basic passage space.

[0073] 2.2 Obstacle verification within the area: Specifically, for each candidate passage area, if there are low obstacles within the candidate passage area whose height is lower than the minimum ground clearance, further verification is required. If the width of a low obstacle is no greater than the width of one tire of the vehicle, and its position is on the expected driving trajectory of one tire of the vehicle, the vehicle can pass over the obstacle by crossing it on one side. If the obstacle is wide or located between the two wheels, it is necessary to determine whether the vehicle can pass directly over the obstacle (i.e., the height of the obstacle is lower than the chassis height, and the vehicle can pass by straddling it; straddling means that the entire chassis of the vehicle passes over the obstacle without the tires on both sides directly contacting it). If there are multiple low obstacles in the candidate passage area, it is necessary to verify whether the vehicle can pass through the entire area by crossing them continuously.

[0074] 2.3 Area Entry / Exit Verification: Specifically, for each candidate passage area, verify whether the lateral width at the entrance and exit of the candidate passage area is greater than the safe passage width for vehicles, ensuring that vehicles can smoothly enter and exit the candidate passage area. If no feasible passage area exists, determine that the road conditions corresponding to the basic parameters of the obstacle do not meet the safe passage conditions, and maintain the original parking decision. If a feasible passage area exists, proceed to the next step.

[0075] 3. Optimal route selection: Specifically, among multiple feasible travel areas (referring to candidate travel areas with feasibility), the optimal travel route is selected. The selection criteria include: Safety first: Prioritize areas with absolutely no obstacles; Minimum number of crossings: Prioritize areas with the fewest low obstacles to cross; Path straightness: Prioritize straight paths that do not require significant turns; Minimize deviation from the original route: Prioritize the area closest to the original driving route.

[0076] Based on the above criteria, each feasible area is comprehensively scored, and the area with the highest score is selected as the final passage route. If there is a final passage route among multiple feasible passage areas, it is determined that the road conditions corresponding to the basic parameters of the obstacle meet the safe passage conditions; otherwise, the original parking decision is maintained.

[0077] 4. Special case handling may include the following situations: Scenario 1: The obstacle forms a "passage" structure; If multiple obstacles form a continuous "passage" structure on both sides of the road (such as continuous obstacles on both sides with a gap in the middle), then according to the logic of the second case, check whether the lateral width of the passage is greater than the safe passage width.

[0078] Scenario 2: The obstacles form an "array" structure; If obstacles are distributed in an array (such as multiple stone blocks arranged in a grid), it is necessary to identify the largest continuous gap in the array and determine whether the gap is passable by a vehicle. The largest continuous gap refers to the largest and most continuous spatial area that can be identified for vehicle passage in a complex distribution of multiple obstacles.

[0079] Scenario 3: No feasible area can be identified; If no suitable passage area can be identified, the vehicle is deemed unable to pass safely, and the original parking decision is maintained.

[0080] It should be noted that all lateral distances mentioned above refer to horizontal distances perpendicular to the vehicle's direction of travel. If obstacles are too densely or complexly distributed, making it impossible to identify a continuous passage path, a degraded approach can be taken: reduce the vehicle speed to the minimum safe speed, or request remote manual intervention for decision-making.

[0081] As an example, when a vehicle enters a parking maneuver due to misidentifying a floating plastic bag as an obstacle, and the user observes that the height of the plastic bag (the vertical distance between its highest point and the ground) is lower than the minimum ground clearance, the user issues a "continue driving" intervention voice command. After receiving the corresponding target voice command, the VCU obtains obstacle parameters from the current driving environment data, including height (vertical distance between the highest point and the ground, e.g., 10cm) and lateral distance (e.g., 4m). It verifies whether the height is lower than the chassis height (minimum ground clearance, e.g., 20cm) and whether the lateral distance is greater than the safe passage width (e.g., 3m). If both conditions are met, the vehicle is deemed to have met the safe passage conditions.

[0082] After determining that the vehicle meets the safe passage conditions, S40 controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions.

[0083] After confirming that safe passage conditions are met, the VCU executes the target voice command and controls the vehicle to perform the corresponding autonomous driving operation. This step realizes the transformation of user intent into actual vehicle control.

[0084] In one specific embodiment, one possible implementation of the above S40 is as follows: S401: Based on the current operating conditions and the target voice command, obtain the target control strategy corresponding to the target voice command; S402, after determining that the vehicle meets the safe passage conditions, controls the vehicle to perform corresponding automatic driving operations according to the target control strategy.

[0085] Among them, the target control strategy is a predefined control logic that describes how the vehicle should respond for specific operating conditions and specific commands. Different combinations of operating conditions and commands correspond to different control strategies.

[0086] The aforementioned target control strategy is designed to enable real-time flexible intervention in vehicle speed adjustment, detours, and other operations, thereby compensating for the lack of flexibility in existing autonomous driving decision-making modules. Users can flexibly intervene in vehicle behavior via voice commands based on their individual needs during the actual sightseeing process, such as slowing down before certain scenic spots or fine-tuning detours according to oncoming traffic. The system executes corresponding operations according to preset strategies and automatically returns the vehicle to its original driving route and normal cruising state after intervention, ensuring the continuity of the driving task.

[0087] Optionally, in this embodiment, multiple control strategies can be pre-constructed based on the mapping relationship between different current operating conditions and different target voice commands, forming a strategy library. Specifically, for each possible current operating condition (such as a parking situation triggered by a suspected obstacle, a constant speed driving situation, a waiting situation for oncoming traffic, or a situation with limited passage space), and for each valid target voice command that may appear under that operating condition (such as a continue driving command, a deceleration command, a resumption of driving command, a detour command, etc.), a corresponding control strategy is pre-configured to clarify the autonomous driving operation that the vehicle should perform.

[0088] In this application's solution, after determining that the vehicle meets safe passage conditions, the system queries and retrieves the target control strategy corresponding to the target voice command from the strategy library based on the current operating conditions and the target voice command. Then, it controls the vehicle to perform corresponding autonomous driving operations according to the target control strategy. This pre-built strategy library approach enables rapid response and precise control of user commands in different scenarios, improving the system's real-time performance and reliability.

[0089] Furthermore, one possible implementation of the above S401 is as follows: S4011, if the current working condition is a working condition that requires intervention due to a suspected obstacle triggering a parking decision, and the target voice command is a command to continue driving, the target control strategy obtained is: cancel the original parking decision and control the vehicle to maintain the preset normal cruising speed and continue driving. S4012, if the current operating condition is constant speed driving condition and the target voice command is a deceleration command, the target control strategy obtained is: control the vehicle to reduce the current speed to the preset low speed range. S4013, if the current operating condition is parking and the target voice command is to resume driving, the target control strategy obtained is: control the vehicle to re-enter the original driving route and restore the preset normal cruise speed. S4014 If the current working condition is waiting for oncoming traffic or limited passage space, and the target voice command is a detour command, the target control strategy obtained is: control the vehicle to complete the detour operation in the direction corresponding to the detour command, and automatically enter the original driving route after detour, and restore the preset normal cruise speed. S4015, if the current operating condition is a non-parking driving condition and the target voice command is a parking command, the obtained target control strategy is: control the vehicle's braking system to activate the parking control unit to perform smooth parking.

[0090] The normal cruising speed refers to the standard driving speed maintained by the vehicle in low-speed sightseeing scenarios such as scenic spots and parks, when it is driving autonomously along a preset route, for example, 8 km / h. The original driving route refers to the predetermined driving trajectory that the vehicle was originally following, determined by a preset path planning algorithm, before the vehicle performs voice-based flexible intervention operations (such as detouring, resuming driving after stopping, etc.). After the flexible intervention is completed (such as completing the detour or resuming driving), the vehicle automatically re-enters the route to ensure the continuity of the driving path and the integrity of the task.

[0091] Smooth parking refers to the process where, when the vehicle is not in a parking position (i.e., during normal driving), it receives a parking command from the user and passes a safety check. Under the control of the vehicle controller, the braking system gradually activates the parking control unit, causing the vehicle to decelerate from its current speed until it comes to a complete stop and remains in a parked state. During this process, the vehicle deceleration is controlled within a preset comfort range (e.g., not exceeding 2.5 m / s²), avoiding discomfort such as vehicle vibration and passenger forward tilting caused by sudden braking, thus achieving smooth and flexible parking control.

[0092] Optionally, see Figure 2 Different control units can be used to achieve autonomous driving control of the vehicle for different target control strategies. Specifically, when the target control strategy is to cancel the stop decision and continue driving, the VCU maintains the current motor speed through the motor control unit to maintain the vehicle speed; when the target control strategy is a deceleration command, the VCU reduces the motor speed through the motor control unit to adjust the vehicle speed to a preset low-speed range; when the target control strategy is a resumption of driving command, the VCU controls the vehicle to re-enter the original driving route through the motor control unit and resumes the preset normal cruising speed; when the target control strategy is a stop command, the VCU activates the braking system through the parking control unit to achieve smooth parking; when the target control strategy is a detour command, the VCU controls the vehicle to fine-tune the driving route according to the command direction through the steering control unit to complete the detour operation, and after detour, the motor control unit and steering control unit work together to make the vehicle automatically resume the original driving route and the preset normal cruising speed. Through this modular control unit configuration, corresponding autonomous driving operations can be precisely executed according to different target control strategies.

[0093] As an example, continuing with the aforementioned obstacle misidentification scenario, after determining that the safe passage conditions are met, the VCU obtains the target control strategy as "cancel the original stop decision and continue driving at the preset normal cruising speed." It then sends a continuous driving signal to the vehicle's power control module, and the vehicle maintains a constant speed of 8 km / h as it passes through the area. Similarly, under constant speed driving conditions, if the user issues a "decelerate" command, the VCU, after verification, obtains the control strategy as "reduce the current speed to a preset low-speed range (e.g., 4-6 km / h)," and then controls the motor to reduce its speed to achieve deceleration. Furthermore, under oncoming traffic waiting conditions, if the user issues a "turn right" command, the VCU, after verification, obtains the control strategy as "complete the detour in the right direction and automatically return to the original driving route and normal cruising speed of 8 km / h after detour," and then controls the steering control unit to veer the vehicle to the right to complete the detour, automatically resetting afterward.

[0094] To enhance the user experience, this method also includes a real-time feedback mechanism. In one specific embodiment, the method further includes: generating feedback information corresponding to the target voice command and / or the target control strategy during the process of controlling the vehicle to perform corresponding autonomous driving operations according to the target control strategy; and providing real-time feedback to the user through voice prompts and / or interface display.

[0095] For example, after receiving and verifying the user's "continue driving" command, the VCU can control the human-machine interaction module to simultaneously issue a voice prompt "command confirmed, continue driving" and display the corresponding information on the interface, so that the user can perceive in real time that the command is received and executed by the system, enhancing the user's sense of control and trust.

[0096] Through the above steps, this disclosure realizes multi-level verification, safety review and scenario-based execution of user voice commands during autonomous driving, effectively solving the technical problems of environmental perception misidentification, lack of flexible intervention methods and insufficient reliability of voice control in the prior art.

[0097] In summary, the technical solution provided in this disclosure addresses the reliability issues of complex environmental noise and the susceptibility of voice commands to mis-triggering in low-speed sightseeing scenarios through a multi-level voice verification mechanism; it resolves the continuity issue of unnecessary stops caused by misidentification of environmental perception modules through safe passage condition verification; and it addresses the flexibility issue of users being unable to intervene in vehicle behavior in real time according to actual needs through a condition-adaptive target control strategy. These mechanisms work synergistically to effectively meet the multiple requirements of low-speed sightseeing scenarios for driving continuity, user interaction flexibility, and voice control reliability, while ensuring the safety of autonomous driving.

[0098] To better illustrate and understand the principles of the method provided in this disclosure, the solution of this disclosure is described below with reference to an optional specific embodiment. It should be noted that the specific implementation of each step in this specific embodiment should not be construed as a limitation of the solution of this disclosure. Other implementation methods that can be conceived by those skilled in the art based on the principles of the solution provided in this disclosure should also be considered within the protection scope of this disclosure.

[0099] In this embodiment, combined with Figure 3 The flowchart further explains the autonomous driving voice-based flexible intervention control method provided in this solution when the current working condition is a waiting-to-stop situation triggered by a suspected obstacle. This method includes: S1, the environmental perception module detects parking, specifically refers to the environmental perception module detecting a suspected obstacle.

[0100] Specifically, when the vehicle is traveling at a preset speed of 8 km / h along the scenic route, the environmental perception module misidentifies a plastic bag floating by the roadside as an obstacle, generates an obstacle detection signal, and sends it to the autonomous driving decision module and the VCU. Based on this signal, the autonomous driving decision module outputs a parking decision command and transmits it to the VCU. The human-machine interaction module simultaneously displays "Obstacle detected, parking is imminent" and issues a voice prompt. At this point, the vehicle enters a parking waiting condition triggered by the suspected obstacle.

[0101] S2, VCU executes the parking request, which specifically refers to the VCU receiving the parking decision instruction and entering the waiting parking state.

[0102] Specifically, after receiving the parking decision instruction from the autonomous driving decision module, the VCU initiates the parking preparation program, controls the power system and braking system to enter standby mode, and continuously obtains current driving environment data from the environmental perception module to prepare for possible user intervention.

[0103] S3 receives the voice module's parsing instructions, specifically referring to receiving the intervention driving voice commands issued by the user after observation and performing hierarchical verification of the intervention driving voice commands.

[0104] Specifically, when passengers inside the vehicle observe that the height of the plastic bag is lower than the vehicle chassis, the system verifies the vehicle's safe passage conditions and issues a voice command to "continue driving." The voice recognition module captures this command through a microphone array, and the adaptive noise reduction unit filters out ambient noise from the scenic area to generate a pre-processed voice signal.

[0105] The VCU control voice command verification module performs four layers of verification: First, noise reduction verification is performed. The signal-to-noise ratio of the preprocessed speech signal is 40dB, which meets the preset signal-to-noise ratio threshold (35dB), and the noise reduction verification is deemed to have passed. Next, keyword matching verification is performed. The voice signal that has passed the noise reduction verification is matched with the "continue driving" command word in the preset keyword library. The matching accuracy (keyword matching degree) reaches 95%, which meets the preset matching degree threshold (90%), and the keyword matching verification is deemed to have passed. Perform the working condition matching verification again to confirm that the "continue driving" command matches the current "waiting to stop working condition" and determine that the working condition matching verification has passed; Finally, permission verification is performed. The voice signal that has passed the working condition matching verification is compared with the preset voice characteristics of the main and passenger personnel. If the permission matching degree reaches 90%, which meets the preset permission threshold (85%), the permission verification is deemed to have passed.

[0106] After all four layers of verification pass, the instruction is identified as the target voice instruction.

[0107] S4, VCU retrieves obstacle data for verification, specifically referring to verifying whether the current driving environment data meets the conditions for safe passage.

[0108] Specifically, the VCU obtains the corresponding safety verification rules based on the command type of the target voice command (continue driving command), retrieves the basic parameters of obstacles in the current driving environment data, such as height 10cm and lateral distance 4m, and performs verification in conjunction with the current operating conditions: If the height of the obstacle is lower than the vehicle chassis height (20cm), the corresponding verification result is "verification passed". If the lateral distance to the obstacle is greater than the safe passage width of the vehicle (3m), the corresponding verification result is "verification passed".

[0109] Both of the above checks passed, indicating that the vehicle meets the safe passage conditions under the current operating conditions.

[0110] S5 refers to the vehicle performing constant-speed automatic driving, which specifically means controlling the vehicle to perform automatic driving operations based on the target voice command and the current operating conditions.

[0111] Specifically, based on the current operating condition (waiting to stop) and the target voice command (continue driving command), the VCU obtains the corresponding target control strategy: cancel the original stopping decision and control the vehicle to continue driving at the preset normal cruising speed.

[0112] The VCU then sends a continuous driving signal to the power control module, canceling the original stop instruction, and the vehicle maintains a constant speed of 8 km / h as it passes through the area. Simultaneously, during the process of controlling the vehicle to perform autonomous driving, the human-machine interface module issues a voice prompt stating "Instruction confirmed, continue driving," which is simultaneously displayed on the interface, providing real-time feedback to the user.

[0113] Through the above S1 to S5, this disclosure successfully realizes multi-level verification of user voice commands, verification of safe passage conditions, and flexible intervention control in the waiting-to-parking situation triggered by suspected obstacles, effectively solving the problem of unnecessary parking caused by misidentification of obstacles.

[0114] The method disclosed herein has the following beneficial effects: First, it significantly improves the accuracy and security of voice control. This disclosure employs a four-layer hierarchical verification mechanism consisting of noise reduction verification, keyword matching verification, operating condition matching verification, and authorization verification. This effectively filters out external noise interference and invalid commands from unauthorized users in complex environments such as scenic spots and parks. Noise reduction verification ensures that the quality of the collected voice signal meets standards; keyword matching verification guarantees accurate recognition of command words; operating condition matching verification verifies the logical consistency between the command and the current scenario; and authorization verification confirms the operator's identity through voiceprint comparison. The synergistic effect of these four layers of verification avoids driving safety hazards caused by noise-induced accidental triggering or unauthorized operations.

[0115] Second, a dual protection mechanism of "user intent + environmental safety" is established. This disclosure introduces a safety passage condition verification based on current driving environment data after the voice command verification is successful, ensuring that the execution of user commands is premised on physical environmental safety. Differentiated safety verification rules are preset for different types of commands. For example, the "continue driving" command requires verification of obstacle height and lateral distance, the "decelerate" command requires verification of whether the target vehicle speed is within a safe range, and the "detour" command requires verification of sufficient lateral space. Through this safety verification mechanism, safety risks caused by user misjudgment or system perception errors are avoided.

[0116] Third, it enables real-time flexible intervention to meet personalized driving needs. This disclosure pre-sets differentiated target control strategies for different operating conditions (waiting to stop, constant speed driving, and limited passage space) and different command types (continue driving, decelerate, resume driving, detour), enabling real-time flexible intervention in operations such as speed adjustment and detour direction. Users can flexibly intervene in vehicle behavior according to actual needs during sightseeing, such as slowing down in front of a scenic spot or fine-tuning the detour direction based on oncoming traffic, thus compensating for the lack of flexibility in existing autonomous driving decision-making modules and significantly improving the user experience.

[0117] Fourth, precise execution and real-time feedback are achieved through modular collaborative control. This disclosure achieves precise execution of different control strategies through modular collaborative control of the parking control unit, motor control unit, and steering control unit. Simultaneously, during the process of controlling the vehicle to perform autonomous driving operations, real-time feedback is provided to the user through voice prompts and interface displays, enabling the user to perceive that commands are received and executed by the system in real time, enhancing the user's sense of control and trust in the vehicle's behavior. This mechanism effectively solves the problem of unnecessary stopping caused by environmental perception misidentification, improving the continuity of the sightseeing process and the riding experience.

[0118] Based on and Figure 1 Using the same principle as the method shown, this disclosure also provides an autonomous driving voice flexible intervention control system 20, such as... Figure 4 As shown, the autonomous driving voice flexible intervention control system 20 may include an acquisition module 210, a verification module 220, an automatic decision-making module 230, and a control module 240, wherein: The acquisition module 210 is used to acquire current driving environment data and user-issued voice commands for intervention during the vehicle's autonomous driving process; the functions of this module are as follows: Figure 2 The system utilizes a microphone array within both the environmental perception module and the voice recognition module. The environmental perception module collects current driving environment data using sensors such as cameras, LiDAR, and millimeter-wave radar; the microphone array captures user-generated voice commands for intervention.

[0119] Verification module 220 is used to perform hierarchical verification of intervention driving voice commands, filtering out external noise and invalid commands from the intervention driving voice commands to obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification, and permission verification. The functions of this module are as follows: Figure 2 The adaptive noise reduction unit and voice command verification module in the speech recognition module work together to achieve this. The adaptive noise reduction unit performs noise reduction processing on the original speech signal; the voice command verification module has a built-in four-layer verification mechanism to perform noise reduction verification, keyword matching verification, working condition matching verification, and permission verification on the noise-reduced voice command, and generate the target voice command.

[0120] The automatic decision-making module 230 is used to determine whether the vehicle meets the preset safe passage conditions under the current operating conditions based on the current driving environment data; the function of this module is as follows: Figure 2 The vehicle control unit (VCU) implements this. The VCU receives current driving environment data from the environmental perception module, combines it with the current operating conditions and the target voice command from the verification module, and executes the logic for judging safe passage conditions.

[0121] The control module 240 is used to control the vehicle to perform autonomous driving based on the target voice command and the current operating conditions after determining that the vehicle meets the safe passage conditions. The functions of this module are as follows: Figure 2 The vehicle control unit (VCU) works in conjunction with the parking control unit, motor control unit, and steering control unit to achieve precise vehicle control. Based on the judgment results and target control strategy, the VCU sends control commands to the corresponding control units: the parking control unit performs parking or parking operations, the motor control unit adjusts the vehicle speed, and the steering control unit controls the driving direction, thereby achieving precise vehicle control.

[0122] Optionally, when the verification module 220 performs hierarchical verification of the intervention driving voice command, filters out external noise and invalid commands in the intervention driving voice command, and obtains the target voice command, it is specifically used for: Determine whether the signal-to-noise ratio (SNR) of the intervention driving voice command is not less than the preset SNR threshold. If the SNR is not less than the SNR threshold, determine that the intervention driving voice command has passed the noise reduction verification. The voice command that has passed the noise reduction verification is matched with the command words in the preset keyword library. The matching degree of the obtained keyword is determined to be no less than the preset matching degree threshold. If the matching degree of the keyword is no less than the matching degree threshold, the voice command that has passed the noise reduction verification is determined to have passed the keyword matching verification. Based on the current working conditions, determine whether the command word corresponding to the voice command verified by keyword matching matches the current working conditions. If the command word matches the current working conditions, determine that the voice command verified by keyword matching has passed the working condition matching verification. The voice commands that have passed the working condition matching verification are matched with the preset voice features of the main and secondary personnel for permission verification. It is determined whether the obtained permission matching degree is not less than the preset permission threshold. If the permission matching degree is not less than the permission threshold, the voice commands that have passed the working condition matching verification are deemed to have passed the permission verification. The voice command that has passed the permission verification is identified as the target voice command.

[0123] Optionally, the aforementioned current driving environment data includes at least one environmental parameter associated with the target voice command. When the automatic decision-making module 230 determines whether the vehicle meets preset safe passage conditions under the current operating conditions based on the current driving environment data, it is specifically used for: Based on the target voice command, obtain the preset security verification rules corresponding to the command type of the target voice command; According to the safety verification rules, at least one environmental parameter in the current driving environment data is verified in conjunction with the current operating condition of the vehicle. If the verification passes, the vehicle is determined to meet the safe passage conditions under the current operating conditions; if the verification fails, the vehicle is determined to not meet the safe passage conditions under the current operating conditions.

[0124] Optionally, before obtaining the intervention driving voice command, the method further includes: The voice intervention module is used to issue a voice intervention prompt when the current driving environment data meets the preset voice intervention trigger conditions. The voice intervention trigger conditions include: the autonomous driving decision module outputs a parking decision, or outputs a detour request, or receives a user's active wake-up command.

[0125] Optionally, when the aforementioned automatic decision-making module 230 verifies at least one environmental parameter in the current driving environment data according to safety verification rules, in conjunction with the vehicle's current operating condition, it is specifically used for: If the current working condition is a parking decision triggered by a suspected obstacle and the target voice command is to continue driving, obtain the basic parameters of the obstacle in the current driving environment data and verify whether the road conditions corresponding to the basic parameters of the obstacle meet the safe passage conditions. If the current driving condition is constant speed driving condition and the target voice command is a deceleration command, obtain the current vehicle speed data in the current driving environment data and verify whether the target vehicle speed after deceleration is within the preset safe vehicle speed range. If the current operating condition is a waiting situation for oncoming traffic or a situation where the passage space is limited, and the target voice command is a detour command, obtain the lateral space data corresponding to the detour direction from the current driving environment data, and verify whether the lateral space data meets the vehicle's safe passage distance requirements.

[0126] Optionally, when the control module 240 determines that the vehicle meets the safe passage conditions and controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions, it is specifically used for: Based on the current operating conditions and the target voice command, obtain the target control strategy corresponding to the target voice command; After determining that the vehicle meets the conditions for safe passage, the system controls the vehicle to perform the corresponding autonomous driving operation according to the target control strategy.

[0127] Optionally, when the control module 240 obtains the target control strategy corresponding to the target voice command based on the current operating conditions and the target voice command, it is specifically used for: If the current working condition is a parking decision triggered by a suspected obstacle and the target voice command is a continue driving command, the target control strategy obtained is: cancel the original parking decision and control the vehicle to maintain the preset normal cruise speed to continue driving. If the current operating condition is constant speed driving and the target voice command is a deceleration command, the target control strategy obtained is: control the vehicle to reduce the current speed to the preset low speed range. If the current operating condition is parking and the target voice command is to resume driving, the target control strategy is to control the vehicle to re-enter the original driving route and restore it to the preset normal cruising speed. If the current working condition is waiting for oncoming traffic or limited passage space, and the target voice command is a detour command, the target control strategy is: control the vehicle to complete the detour operation in the direction corresponding to the detour command, and automatically return to the original driving route after detour and restore the preset normal cruise speed. If the current operating condition is a non-parking driving condition and the target voice command is a parking command, the obtained target control strategy is: control the vehicle's braking system to activate the parking control unit to perform smooth parking.

[0128] Optionally, the system also includes: The interaction module is used to generate feedback information corresponding to the target voice commands and / or target control strategies during the process of controlling the vehicle to perform corresponding autonomous driving operations according to the target control strategy; and to provide real-time feedback to the user through voice prompts and / or interface display.

[0129] The autonomous driving voice flexible intervention control system of this disclosure can execute the autonomous driving voice flexible intervention control method provided in this disclosure. The implementation principle is similar. The actions performed by each module and unit in the autonomous driving voice flexible intervention control system in each embodiment of this disclosure are corresponding to the steps in the autonomous driving voice flexible intervention control method in each embodiment of this disclosure. For detailed functional descriptions of each module of the autonomous driving voice flexible intervention control system, please refer to the descriptions in the corresponding autonomous driving voice flexible intervention control methods shown above. They will not be repeated here.

[0130] The aforementioned autonomous driving voice flexible intervention control system can be a computer program (including program code) running on a computer device, such as an application software; the system can be used to execute the corresponding steps in the methods provided in the embodiments of this disclosure.

[0131] In some embodiments, the autonomous driving voice flexible intervention control system provided in this disclosure can be implemented in a combination of hardware and software. As an example, the autonomous driving voice flexible intervention control system provided in this disclosure can be a processor in the form of a hardware decoding processor, which is programmed to execute the autonomous driving voice flexible intervention control method provided in this disclosure. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0132] In other embodiments, the autonomous driving voice-based flexible intervention control system provided in this disclosure can be implemented in software. Figure 4 An autonomous driving voice-based flexible intervention control system stored in a memory is shown. It can be software in the form of programs and plug-ins, and includes a series of modules, including an acquisition module 210, a verification module 220, an automatic decision-making module 230, and a control module 240, for implementing the autonomous driving voice-based flexible intervention control method provided in the embodiments of this disclosure.

[0133] The modules described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules are not, in some cases, intended to limit the functionality of the module itself.

[0134] Based on the same principles as the systems shown in the embodiments of this disclosure, embodiments of this disclosure also provide a vehicle that includes the aforementioned autonomous driving voice flexible intervention control system.

[0135] Based on the same principles as the methods shown in the embodiments of this disclosure, the embodiments of this disclosure also provide an electronic device, which may include, but is not limited to: a processor and a memory; the memory for storing computer programs; and the processor for executing the methods shown in any embodiment of this disclosure by invoking the computer programs.

[0136] In one alternative embodiment, an electronic device is provided, such as Figure 5 As shown, Figure 5The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of this disclosure.

[0137] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with this disclosure. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0138] Bus 4002 may include a pathway for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0139] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0140] The memory 4003 stores application code (computer program) that executes the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0141] Among these, electronic devices can also be terminal devices. Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0142] This disclosure provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0143] According to another aspect of this disclosure, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various embodiments described above.

[0144] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0145] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0146] The computer-readable storage medium provided in this disclosure can be, for example,—but is not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device.

[0147] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0148] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A method for flexible voice intervention control in autonomous driving, characterized in that, For applications in vehicles that include a vehicle controller, the following steps are included: During the autonomous driving process, data on the current driving environment under the current operating conditions and the user's voice commands for intervention are acquired. The intervention driving voice command is subjected to hierarchical verification to filter out external noise and invalid commands in the intervention driving voice command to obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. Based on the current driving environment data, determine whether the vehicle meets the preset safe passage conditions under the current operating conditions; After determining that the vehicle meets the safe passage conditions, the system controls the vehicle to perform autonomous driving based on the target voice command and the current operating conditions.

2. The method according to claim 1, characterized in that, The hierarchical verification of the intervention driving voice command, filtering out external noise and invalid commands from the intervention driving voice command to obtain the target voice command, includes: Determine whether the signal-to-noise ratio (SNR) of the intervention driving voice command is not less than a preset SNR threshold. If the SNR is not less than the SNR threshold, determine that the intervention driving voice command has passed the noise reduction verification. The voice command that has passed the noise reduction verification is matched with the command words in the preset keyword library. It is determined whether the obtained keyword matching degree is not less than the preset matching degree threshold. If the keyword matching degree is not less than the matching degree threshold, it is determined that the voice command that has passed the noise reduction verification has passed the keyword matching verification. Based on the current working condition, determine whether the instruction word corresponding to the voice instruction verified by keyword matching matches the current working condition. If the instruction word matches the current working condition, determine that the voice instruction verified by keyword matching has passed the working condition matching verification. The voice command that has passed the working condition matching verification is matched with the preset voice features of the main and passenger personnel for permission verification. It is determined whether the obtained permission matching degree is not less than the preset permission threshold. If the permission matching degree is not less than the permission threshold, it is determined that the voice command that has passed the working condition matching verification has passed the permission verification. The voice command that has passed the permission verification is identified as the target voice command.

3. The method according to claim 1, characterized in that, Before obtaining the intervention driving voice command, the process also includes: In response to the current driving environment data meeting the preset voice intervention trigger conditions, a voice intervention prompt is issued. The voice intervention trigger conditions include: the autonomous driving decision module outputs a parking decision, or outputs a detour request, or receives a user's active wake-up command.

4. The method according to claim 1, characterized in that, The current driving environment data includes at least one environmental parameter associated with the target voice command. The step of determining whether the vehicle meets preset safe passage conditions under the current operating conditions based on the current driving environment data includes: Based on the target voice command, obtain the preset security verification rules corresponding to the command type of the target voice command; According to the safety verification rules, at least one environmental parameter in the current driving environment data is verified in conjunction with the current operating condition of the vehicle. If the verification passes, the vehicle is determined to meet the safe passage conditions under the current operating conditions; if the verification fails, the vehicle is determined to not meet the safe passage conditions under the current operating conditions.

5. The method according to claim 4, characterized in that, The step of verifying at least one environmental parameter in the current driving environment data according to the safety verification rules, in conjunction with the current operating condition of the vehicle, includes: When the current working condition is a working condition that requires intervention due to a suspected obstacle triggering a parking decision, and the target voice command is a command to continue driving, the basic parameters of the obstacle in the current driving environment data are obtained, and the road conditions corresponding to the basic parameters of the obstacle are verified to meet the safe passage conditions. When the current operating condition is constant speed driving and the target voice command is a deceleration command, the current vehicle speed data in the current driving environment data is obtained, and it is verified whether the target vehicle speed after deceleration is within the preset safe vehicle speed range. When the current operating condition is a waiting condition for oncoming traffic or a condition with limited passage space, and the target voice command is a detour command, the lateral space data corresponding to the detour direction in the current driving environment data is obtained, and the lateral space data is verified to meet the safe passage distance requirements of the vehicle.

6. The method according to any one of claims 1 to 5, characterized in that, After verifying that the vehicle meets the safe passage conditions, the step of controlling the vehicle to perform autonomous driving based on the target voice command and the current operating conditions includes: Based on the current operating conditions and the target voice command, obtain the target control strategy corresponding to the target voice command; After determining that the vehicle meets the safe passage conditions, the vehicle is controlled to perform corresponding autonomous driving operations according to the target control strategy.

7. The method according to claim 6, characterized in that, The step of obtaining the target control strategy corresponding to the target voice command based on the current operating condition and the target voice command includes: When the current operating condition is a stop decision triggered by a suspected obstacle, and the target voice command is a continue driving command, the acquired target control strategy is: cancel the original stop decision and control the vehicle to maintain its position. Preset normal cruise The vehicle continued to travel at the specified speed; When the current operating condition is constant speed driving and the target voice command is a deceleration command, the obtained target control strategy is: control the vehicle to reduce the current speed to a preset low speed range. When the current operating condition is parking and the target voice command is a command to resume driving, the target control strategy obtained is: to control the vehicle to re-enter the original driving route and restore it to the preset normal cruising speed. When the current working condition is a waiting for oncoming traffic or a limited passage space, and the target voice command is a detour command, the target control strategy obtained is: control the vehicle to complete the detour operation in the direction corresponding to the detour command, and automatically enter the original driving route after detour, and restore to the preset normal cruising speed. When the current operating condition is a non-parking driving condition and the target voice command is a parking command, the obtained target control strategy is: control the vehicle's braking system to activate the parking control unit to perform smooth parking.

8. The method according to claim 6, characterized in that, The method further includes: During the process of controlling the vehicle to perform corresponding autonomous driving operations according to the target control strategy, feedback information corresponding to the target voice command and / or the target control strategy is generated. The feedback information will be provided to the user in real time through voice prompts and / or interface display.

9. An autonomous driving voice-based flexible intervention control system, characterized in that, Applicable to vehicles that include a vehicle controller, including: The acquisition module is used to acquire current driving environment data and user-issued voice commands for intervention during the autonomous driving process. The verification module is used to perform hierarchical verification of the intervention driving voice command, filter out external noise and invalid commands in the intervention driving voice command, and obtain the target voice command. The hierarchical verification includes at least noise reduction verification, keyword matching verification, working condition matching verification and permission verification. An automatic decision-making module is used to determine whether the vehicle meets preset safe passage conditions under the current operating conditions based on the current driving environment data. The control module is used to control the vehicle to perform autonomous driving based on the target voice command and the current operating conditions after determining that the vehicle meets the safe passage conditions.

10. A vehicle, characterized in that, Including the autonomous driving voice flexible intervention control system as described in claim 9.

Images