Multi-mode voice control tail gate action management method based on vehicle state recognition

Abstract

The invention relates to the technical field of tail gate control, in particular to a multi-mode voice control tail gate action management method based on vehicle state recognition, which comprises the following steps of: acquiring an ACC level, a door lock state and a tail gate closing signal, recording an ACC interruption time period, classifying into a static vehicle scene by combining the door lock state and the tail gate state, and performing voice control. And collecting a foot track extraction direction vector and matching a cursor area, extracting a voice frame segment and positioning an overlapping section, selecting a main control channel to access a tail gate interface, and inputting an instruction to a controller to obtain a tail gate execution state identifier. According to the method, a static scene triggering condition is constructed through signal cross recognition, the control judgment precision is improved, the foot action analysis capacity is enhanced through track direction offset matching, dynamic linkage recognition is carried out through voice and track overlapping positioning, the input error-tolerant rate and the anti-interference capacity are improved, and instruction response logic is optimized through channel sequence setting; limiting feedback is combined with displacement path matching, the motion of the tail gate is kept complete, and a high-stability control process is formed.

Description

Technical Field

[0001] The present invention relates to the technical field of tailgate control, and in particular to a multimodal voice-controlled tailgate action management method based on vehicle state recognition. Background Art

[0002] The field of tailgate control technology encompasses the management of the opening and closing of the vehicle's rear enclosure and the design of control strategies. Core aspects of this technology include the tailgate's motion mechanism, control signal recognition, linkage with the vehicle's overall systems, and control logic based on user interaction. Tailgate control technology encompasses the drive control of the mechanical actuator, the command response logic of the electronic control unit, and the analysis of user operation (e.g., buttons, capacitive sensing, voice commands, etc.). Development trends in this field include multimodal input fusion, optimized control response accuracy, and deep integration with the vehicle's intelligent systems.

[0003] The multimodal voice-controlled tailgate action management method based on vehicle state recognition manages tailgate opening and closing movements during the tailgate control process, combining the vehicle's current state information with control commands generated through a combination of voice and other input methods. This method addresses technical issues including vehicle state acquisition and recognition, voice input parsing, and control command generation and issuance. Specifically, tailgate action authority is determined based on vehicle speed information, gear position, parking status, and other factors. This is supplemented by the combined analysis of voice input commands and touch interaction data to generate tailgate opening and closing commands, which are then executed by the control system on the tailgate drive mechanism.

[0004] Control authority determination relies on a single state signal, such as detecting parking or speed alone, while ignoring other critical signal interactions. This results in insufficient control scenario recognition, making it prone to misjudging the vehicle's critical state and causing control authority to be inadvertently unlocked or locked. Foot recognition relies primarily on motion triggering events rather than path characteristics, lacking a mechanism to model the spatial path of the motion process and identify directional differences. This makes it unable to accurately reflect motion deviations caused by individual differences. Voice control lacks synergy with other input methods, making audio input prone to false triggering or invalid recognition in non-quiet environments, and thus lacking robust recognition. A lack of interaction order recognition and control priority processing between multiple input paths can easily lead to response confusion and control logic imbalance when voice and motion are concurrent. Simple limit determination is often used during action execution, failing to integrate real-time path feedback for dynamic matching, resulting in response delays or the risk of misalignment. These issues limit the control performance of existing technologies under complex interaction conditions, impacting the accuracy and stability of tailgate control systems. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a multimodal voice-controlled tailgate action management method based on vehicle state recognition.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a multimodal voice-controlled tailgate action management method based on vehicle state recognition, comprising the following steps: S1: Acquires the ACC level, door lock status, and tailgate closing signal, records the ACC interruption period, identifies the door lock unlocking status, and combines it with the tailgate closing feedback to process the intersection of the signal states and classify them into a stationary vehicle scenario to obtain the tailgate management activation status indicator; S2: Based on the tailgate management activation state flag, the foot trajectory is collected and the path start and end points and direction vectors are extracted. The trajectory convergence direction is matched with the offset difference and compared with the cursor area to obtain the kick control trigger instruction flag; S3: Based on the kick control trigger instruction identifier, extract the voice instruction parsing frame segment and mark the recognition status, retrieve the overlapping segment of the kick path and the voice frame sequence, locate the concurrent interaction state, and obtain the multimodal control intervention state identifier; S4: Based on the multimodal control intervention state identifier, extract the input frame sequence, select the channel of the first input structure as the main control path, inject it into the tailgate control interface, and obtain the tailgate main control channel setting state; S5: Based on the setting state of the tailgate master control channel, the command frame sequence and the static flag are input, and the corresponding displacement process path is fed back through the limit to obtain the tailgate action management execution state identifier.

[0007] As a further solution of the present invention, the tailgate management activation status identifier includes the ACC interrupt holding time, level stability state parameters, door lock unlocking level type, tailgate closing feedback signal type, and signal combination classification flag; the kick control trigger instruction identifier includes the trajectory start and end time, Z-axis path change vector, convergence offset matching result, cursor overlap area number, and closed path annotation status; the multimodal control intervention state identifier includes the kick path number, voice frame segment number, voice active label, time overlap segment number, and interactive concurrency identification flag; the tailgate main control channel setting state includes the main control channel number, the first segment complete structure mark, the path continuity mark, the input sequence positioning label, and the control access channel identification code; the tailgate action management execution state identifier includes the input control frame segment number, the static state matching flag, the tailgate action feedback segment displacement sequence, the feedback closure confirmation signal, and the controller action execution status code.

[0008] As a further solution of the present invention, the specific steps of S1 are: S101: Acquire the vehicle's current ACC level status, door lock status, and tailgate closed status, collect interruption points in the ACC level signal and mark the time period during which it is continuously maintained, read the unlocked state level signal type from the door lock control line, extract the closed feedback signal from the tailgate limit switch, and obtain a state signal acquisition record; S102: Based on the state signal collection record, extracting a signal combination of ACC hold interruption, door lock unlocking, and electric tailgate closing from the ACC level, door lock signal, and tailgate feedback, and comparing the state consistency of the signals within the same time period to obtain a static signal combination state; S103: Based on the static condition signal combination state, compare the static state trigger condition in the tailgate control scene, match the signal combination with the trigger condition, access the control entrance, set it to the tailgate control preparation state, and obtain the tailgate management activation state identifier.

[0009] As a further solution of the present invention, the specific steps of S2 are: S201: Based on the tailgate management activation state identification period, Z-axis displacement values and corresponding timestamps within the continuous foot movement area are collected, the start and end time periods of the path are extracted, and the direction vectors between each frame are marked to obtain a foot path direction dataset; S202: Based on the foot path direction dataset, extract the starting direction and ending direction of the path, calculate the angle between the starting and ending points of the path, compare the angle with a convergence judgment angle reference value, identify the path shape that tends to converge in a single direction, and obtain a path convergence matching value; S203: Based on the path convergence matching value, the trajectory boundary coordinate points are extracted and compared with the boundary range of the projection area, and the interactive path of the trajectory forming a closed structure within the boundary range is identified to obtain the kick control trigger instruction identifier.

[0010] As a further solution of the present invention, the calculation formula of the angle between the starting and ending points of the path is specifically: ; in, Represents the path starting and ending point angle between the direction vectors of the path starting point and the end point, Represents the direction vector of the starting point of the path, Represents the direction vector of the path end point, Represents the sum of the displacement increments of the segments in the path in the x-axis direction, Represents the average value of the displacement increment of the segments in the path in the y-axis direction, Represents the standard deviation of the displacement increment in the x-axis direction of the path, Represents the standard deviation of the displacement increment in the y-axis direction of the path.

[0011] As a further solution of the present invention, the specific steps of S3 are: S301: Based on the kick control trigger instruction identifier, extract the instruction parsing frame segment in the current speech recognition channel, record the start frame and end frame numbers of the speech input, and mark the active tag of the current recognition state to obtain speech recognition frame segment state information; S302: Based on the speech recognition frame segment state information, according to the start and end frame numbers of the kick path, the speech and kick frame segments are time-correlated, the intersection segment between the start and end frames is extracted, and the number of overlapping frames in the intersection segment is calculated to obtain the number of overlapping frames of the input frame segment; S303: Based on the number of overlapping frames of the input frame segments, extract the frame segments as the interaction basis for the overlap of the voice input and the kicking path in time sequence, set the frame segment status to the concurrent input mark content, and obtain the multimodal control intervention state identifier.

[0012] As a further solution of the present invention, the calculation formula for the number of overlapping frames in the intersection interval is specifically: ; in, Represents the number of overlapping frames in the crossover interval, Represents each frame number within the crossover interval, Represents the start frame of the cross section, Represents the end frame number of the crossover segment, Representative The timestamp of the frame kick line, Representative Frame speech timestamp, Representative kick The path direction offset value of the frame, Representative voice The semantic parsing strength value of the frame, Represents the square value of the propagation stability of the path frame on the time axis, Represents the square value of the retention time of the speech recognition frame on the time axis.

[0013] As a further solution of the present invention, the specific steps of S4 are: S401: Based on the multimodal control intervention state identifier, extract the start and end frame numbers of the kick and voice channels, arrange the channel input sequences in frame segment order, and verify the channel source paths of the two groups of inputs in combination with the tag information to obtain the channel input sequence structure value; S402: Based on the channel input sequence structure value, extract the start frame number and the end frame number of the two channels, calculate the coverage span value between the first frame and the last frame of each channel, and extract the frame segment with complete coverage range as the candidate path according to the span value to obtain the channel frame segment coverage span S403: Based on the channel frame segment coverage span, the channel corresponding to the coverage span value is set as the current main control path, and the input identification information corresponding to the channel is connected to the tailgate control interface to obtain the tailgate main control channel setting state.

[0014] As a further solution of the present invention, the calculation formula for the coverage span value between the first frame and the last frame of each channel is specifically: ; in, Representative Channel The coverage span value between the first frame and the last frame of Representative Channel The ending frame number, Representative Channel The starting frame number, Representative Channel The number of frame-segment pairs involved in the local structure weight analysis, Representative Channel No. The end frame number of the local frame segment, Representative Channel No. The starting frame number of a local frame segment, Representative Channel The average value of the start frame numbers of all local frame segments, Representative Channel The average value of the end frame number of the local frame segment, Representative Channel No. The weighting factor of each local frame segment.

[0015] As a further solution of the present invention, the specific steps of S5 are: S501: Based on the control source recorded in the tailgate master control channel setting state, the frame sequence number of the channel's current input control command is collected, and the static flag signal in the vehicle state is synchronously extracted and uniformly packaged into a path signal package to obtain a control input and state synchronization data set; S502: Based on the control input and state synchronization data set, extract the closing feedback state of the tailgate stopper and the displacement feedback sequence during the tailgate action execution process, and perform frame-segment sequential comparison between the displacement target value in the input control frame sequence and the feedback path to obtain control path correspondence information; S503: Based on the control path correspondence information, the signal matching the control frame sequence structure in the feedback path is connected to the input port of the tailgate execution controller, the encapsulated input frame structure is identified, and the tailgate action process is scheduled after the control path is written to obtain the tailgate action management execution status identifier.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, the trigger conditions for static scenes are constructed through signal cross-recognition to improve the control judgment accuracy, the trajectory direction offset matching enhances the ability to analyze foot movements, the voice and trajectory overlap positioning are used for dynamic linkage recognition to improve the input fault tolerance and anti-interference ability, the channel sequence setting optimizes the command response logic, the limit feedback is combined with the displacement path matching to keep the tailgate action intact, and form a highly stable control process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the main steps of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0020] See also Figure 1 The present invention provides a technical solution: a multimodal voice-controlled tailgate action management method based on vehicle state recognition, comprising the following steps: S1: Obtain the vehicle's current ACC level status, door lock status, and tailgate open / close status. ACC signal interruption points are extracted and recorded for duration. The unlock level status on the door lock control line is collected. Status comparisons are performed on the vehicle's stability during standby. Combined with the tailgate limiter feedback, the signal statuses are intersected and classified into a stationary vehicle scenario to obtain the tailgate management activation status indicator. S2: During the tailgate management activation status identification period, light-sensing path tracking is used to collect continuous foot motion trajectories and extract the time start and end ranges and path change vectors. The trajectory convergence direction is matched by offset difference. The trajectory range is cross-compared with the cursor area position, and the interaction process of the completed closed trajectory is recorded to obtain the kick control trigger instruction identification. S3: Based on the kick control trigger command identifier, the voice command parsing frame segments are synchronously extracted and the current recognition state is marked as active. The start and end time periods of the kicking behavior path and the voice recognition frame sequence are searched for overlapping areas. The time range where the intersection exists is located as a concurrent interaction state input situation, and the multimodal control intervention state identifier is obtained. S4: Based on the concurrency of multimodal control intervention state identification, the order of the kick and voice channel input frames is extracted. The input sequence and channel command completeness flag in the path source tag are called to identify the first command source with a complete input structure and continuous frame parsing characteristics. This is marked as the current main control path and then connected to the tailgate control interface to obtain the tailgate main control channel setting status. S5: Based on the control source recorded in the tailgate master control channel setting status, the current tailgate input control command frame sequence and the vehicle status static flag are synchronously transmitted to the execution controller. The path correspondence processing is performed through the tailgate limit feedback status and the displacement process in the action execution feedback. The controller opens and closes the tailgate according to the relationship between the input command and the feedback status, and obtains the tailgate action management execution status identifier.

[0021] The tailgate management activation status identifier includes the ACC interrupt retention time, level stability state parameters, door lock unlocking level type, tailgate closing feedback signal type, and signal combination classification identifier. The kick control trigger instruction identifier includes the trajectory start and end time, Z-axis path change vector, convergence offset matching result, cursor overlap area number, and closed path annotation status. The multimodal control intervention status identifier includes the kick path number, voice frame segment number, voice active label, time overlap segment number, and interactive concurrency identification flag. The tailgate master control channel setting status includes the master control channel number, first segment complete structure mark, path continuity identifier, input sequence positioning label, and control access channel identification code. The tailgate action management execution status identifier includes the input control frame segment number, static state matching flag, tailgate action feedback segment displacement sequence, feedback closure confirmation signal, and controller action execution status code.

[0022] The specific steps of S1 are: S101: Acquire the vehicle's current ACC level status, door lock status, and tailgate closed status, collect interruption points in the ACC level signal and mark the time period during which it is continuously maintained, read the unlocked state level signal type from the door lock control line, extract the closed feedback signal from the tailgate limit switch, and obtain a state signal acquisition record; First, the corresponding signal data is collected from the vehicle power controller, door lock control circuit and tailgate limiter in turn. The extraction of ACC level requires continuous sampling of the ignition power supply voltage output signal, and the detection cycle is set to 500ms. The continuous state of voltage higher than 11.5V is defined as ACC on, and the corresponding interruption point is the time point when the voltage is converted from the on state to the state below 1V. The time period after this point is marked on the time axis. For example, if the ACC voltage of a vehicle drops sharply from 12.1V to 0.8V between 08:30:00 and 08:30:05 and remains for more than 5 seconds, this time period is marked as the interruption section; the door lock status acquisition requires reading the level type of the unlocking signal in the door control line. The unlocking state is usually output as 12V DC, which needs to be maintained for more than 5 frames in the frame sequence sampling, and the interval time between each frame is not less than A valid unlock state can only be confirmed after 200ms. If the level of 12V is continuously recorded for five frames between 08:31:12 and 08:31:13, it is considered to be in a stable unlock state. The tailgate closed state is composed of the feedback signal of the mechanical contact in the limiter. The normally closed contact is open, indicating that the tailgate is open, and the output is 5V when closed. The acquisition process needs to read the contact closure status between two time points and record the closed frame segment. For example, if the feedback is 5V continuously for 10 frames from 08:32:10 to 08:32:12, it can be determined that the tailgate is closed. After the three types of data are collected, they should be combined into a multi-signal sequence according to the unified timestamp alignment format. The frame sequence record matrix is generated in chronological order. The acquisition source, signal consistency and time synchronization are verified frame by frame. Finally, the state linkage flag is established as a prerequisite for the subsequent tailgate control trigger, and the state signal acquisition record is obtained.

[0023] S102: Based on the state signal collection and recording, extract the signal combination of ACC hold interruption, door lock unlocking, and electric tailgate closing from the ACC level, door lock signal, and tailgate feedback. Compare the state consistency of the signals within the same time period to obtain the static signal combination state. First, based on the frame sequence data in the status signal acquisition record, the start and end time points of the ACC level interruption segment are extracted one by one. The frame number range of the door lock level state in the unlocked level output is read. At the same time, the stable feedback segment corresponding to the tailgate limit signal in the closed state is extracted. Each frame of the three signals contains fields including: signal name, start time, end time, and status value. During the extraction stage, it is necessary to determine whether the status values of the three types of signals are stable. If the ACC signal voltage remains in the 0V range for more than 60 seconds, the door lock signal remains at a level of 12V in the unlocked output range for more than 30 seconds, and the tailgate closed signal remains at a low level of 0V without voltage jumps for more than 20 milliseconds, then the three signals are considered stable. Based on this, the three extracted time intervals are respectively subjected to time period intersection processing. That is, based on the frame timestamp, the minimum common overlapping segment in which the three signal states are simultaneously established is found. For example, the ACC level signal is interrupted from 10:01:00 to 10:03:00, and the door lock signal is resolved from 10:01:30 to 10:04:00. Lock state, tailgate closing signal feedback time is 10:00:50 to 10:02:50, then the three overlap period is 10:01:30 to 10:02:50. In this overlap period, each signal state is judged one by one with the state acquisition reference value. If the state value of each frame in the corresponding judgment is the target stable state, a valid signal combination is formed. This signal combination is coded and identified to form a signal intersection matching record, and written into the downstream judgment structure for subsequent processing node call. In the above processing process, in order to judge Signal state consistency also requires determining whether the state values in the three frame types meet the requirements of synchronization at the same time node. For example, at 10:01:45, if the ACC is OFF, the door locks are unlocked, and the tailgate is closed, then all three judgments are true, and the signal valid frame is recorded as 1. If any signal in a frame does not meet the above state values, the current frame is skipped. The valid combination frame segment length and its start and end times are determined by frame-by-frame screening. Finally, the continuous frame segments that meet the judgment conditions are defined as the signal consistency interval, and the static signal combination state is obtained.

[0024] S103: Based on the static condition signal combination state, the static state trigger condition in the tailgate control scenario is compared, the signal combination is matched with the trigger condition, a control entry is accessed, the tailgate control preparation state is set, and a tailgate management activation state flag is obtained; First, we need to clarify which signals the static condition signal combination state mentioned above consists of. These signals usually include the ACC level being in the disconnected state, the power supply voltage being lower than 1V for more than 2 seconds, the door lock being in the unlocked state, and the tailgate limiter feedback being in the closed state for 10 consecutive frames. These signals must be met at the same time and overlap within the same time sequence. They are locked as a combined input signal set. At this time, the static trigger condition set in the vehicle tailgate control process needs to be called. The static condition is generally composed of the logic level state values corresponding to each signal. For example: ACC is low, the door lock is high, the tailgate limiter feedback is high and the door control action signal is not triggered, and they are all within the trigger judgment window time period. This time period is generally set to 3 seconds. In actual vehicle applications, this can be judged by an example: if at 08:42: During the time period from 00:00 to 08:42:03, the ACC voltage is stable at 0.8V, the door lock voltage is 12.2V, the limit feedback is 5V, and there is no control command signal flowing in. This matches the static scene condition. Subsequently, the collected signal frame group needs to be compared with the status value in the preset trigger condition in frame sequence. In each frame, if the status of all parameters is the same as the standard value, it is recorded as a valid match frame. The number of consecutive hit frames of the frame group must reach no less than 5 frames to be considered a stable state match. After the comparison is completed, the valid matching signal combination is written into the tailgate control command entry data stream as a control condition registration behavior. After reading it, the control entry marks the identification signal status frame group with a "ready state" label, completes the pre-activation preparation process of the control channel, and obtains the tailgate management activation state identification.

[0025] The specific steps of S2 are: S201: Based on the tailgate management activation state identification period, Z-axis displacement values and corresponding timestamps within the continuous foot movement area are collected, the start and end time periods of the path are extracted, and the direction vectors between each frame are marked to obtain a foot path direction dataset; While the vehicle remains stationary, the spatial change trajectory of the rear foot interaction area is continuously monitored. The kicking action is collected through ToF or infrared depth perception sensors. The sensor samples the vertical distance information of the target object in the preset area at a frequency of 10 frames per second. The Z-axis displacement value is obtained with the ground as the 0 reference. If the depth value of the target object suddenly changes from 150mm to 800mm and remains in this range for more than 5 frames, it is determined that there is an initial behavior of the foot entering the area. The timestamp is recorded as the system millisecond time when the distance mutation is detected in the first frame, such as 08:35:12.253. This time point is used as the path starting time frame. Similarly, when the target object leaves the area after continuous displacement or the displacement difference is less than 10mm for more than 5 consecutive frames, the system determines the path end frame, such as 08:35:13.089, then the path start and end time period. The time interval is 836 milliseconds. The direction of the spatial position change between each frame is further determined. The direction vector is calculated according to the coordinate difference vector of each two frames. If the position of the t1 frame is (0, 0, 300mm) and the t2 frame is (10mm, 5mm, 600mm), the direction vector is ΔX=10, ΔY=5, ΔZ=300. The direction vector calculation is used to subsequently determine whether the foot path has aggregation, forward or concentration characteristics. If the direction of continuous direction vector changes is consistent, it means that the foot behavior has complete flow directionality. If the direction fluctuates frequently, it means that there may be mistouch behavior. The direction vector between each frame needs to be directly calculated based on the spatial three-axis difference, and finally compared and analyzed with the spatial behavior feature library required for kicking action judgment. After completing displacement extraction, time positioning, and direction labeling, the continuous path segment is uniformly imported into the data record structure to obtain the foot path direction dataset.

[0026] S202: Based on the foot path direction dataset, extract the starting direction and ending direction of the path, calculate the angle between the starting and ending points of the path, compare the angle with the convergence judgment angle reference value, identify the path shape that tends to converge in a single direction, and obtain the path convergence matching value; The calculation formula for the angle between the starting and ending points of the path is: ; in, Represents the path starting and ending point angle between the direction vectors of the path starting point and the end point, Represents the direction vector of the starting point of the path, Represents the direction vector of the path end point, Represents the sum of the displacement increments of the segments in the path in the x-axis direction, Represents the average value of the displacement increment of the segments in the path in the y-axis direction, Represents the standard deviation of the displacement increment in the x-axis direction of the path, Represents the standard deviation of the displacement increment in the y-axis direction of the path; Assumption: The first point in the path is (1.2, 3.1); The next point is (2.5, 4.8); get: ; The corresponding module length is: ; The last point of the path is (6.7, 8.9), and the end point is (7.6, 10.2), so: ; The corresponding module length is: ; Vector dot product: ; The path contains a total of 8 segment points. According to the trajectory coordinate sequence, the displacement increments in the x-axis direction are obtained through trajectory processing statistics: 1.3, 0.9, 1.1, 1.0, 1.4, 1.2, 1.3, 1.1.

[0027] Obtain: ; The displacement increments in the y-axis direction are: 1.7, 1.3, 1.5, 1.2, 1.6, 1.4, 1.3, 1.2. Calculate the average value: ; The standard deviation in the x-axis direction is: ; ; ; ; The standard deviation in the y-axis direction is calculated in the same way. ,have to: ; Substituting the above values into the formula: ; calculate: ; ; ; Substitute: ; The result shows that the angle tends to the minimum value and the path direction is highly consistent, indicating that the path has strong directional convergence and meets the single-direction path convergence feature matching condition. The final result is compared by the angle threshold (such as 20 degrees), and the path convergence matching value is obtained as convergence is established.

[0028] S203: Based on the path convergence matching value, the coordinate points of the trajectory boundary are extracted and compared with the boundary range of the projection area to identify the interactive path of the trajectory forming a closed structure within the boundary range, and obtain the kick control trigger instruction identifier; First, the spatial coordinate points corresponding to the outer edge of the path are extracted from the calibrated foot behavior trajectory data. The identification of the boundary coordinate points needs to be based on the maximum outer expansion range of the trajectory. The trajectory envelope is located by the set of XYZ three axial coordinate points in each frame path, and the position points corresponding to the first frame, the last frame, and the maximum offset frame are extracted to construct the trajectory boundary component surface. For example, in the trajectory frame segment, the coordinates of the first frame are (0, 0, 200), the last frame is (60, 20, 650), and the maximum offset frame is (45, 65, 430). The boundary range can be generated by the three-dimensional area determined by these three points, and then compared with the boundary range of the light-sensing projection area set at the rear of the vehicle. The boundary range can be formed by a preset infrared comparison light curtain, and its boundary is covered by a rectangular four-point area, for example, the coordinates are 0 to 100 in the X direction, 0 to 100 in the Y direction, and 200 to 700 in the Z direction. On this basis, By comparing the projections of the trajectory boundary points and the light-sensing area boundary points in the three axes, it is determined whether all boundary points fall within the coordinate range covered by the projection area. If the coordinates of all boundary points are within the projection area, the trajectory is considered to be a path behavior that matches the projection area in space. The proximity of the starting point and the end point in the closed path under the spatial projection is further evaluated. If the spatial straight-line distance between the starting point and the end point is less than the set convergence threshold, for example, less than 50mm, the path meets the closed-loop feature. In practical applications, for example, when the first and last points of the path are (5, 5, 200) and (8, 6, 195), the distance between the two points is about 6.7mm, which meets the convergence closure requirements. After meeting the dual conditions of boundary coverage and head-to-tail convergence, the trajectory can be identified as a complete and valid kicking interaction behavior, and the kick control trigger instruction identifier is obtained.

[0029] The specific steps of S3 are: S301: Based on the kick control trigger command identifier, extract the command parsing frame segment in the current speech recognition channel, record the start and end frame numbers of the speech input, and mark the active label of the current recognition state to obtain speech recognition frame segment state information; Extracting the command parsing frame segments in the current voice recognition channel requires taking the effective confirmation time point of the kick trigger as the benchmark, locating the content currently in the monitoring state in the voice input data stream, and extracting the frame segments within the acquisition window at a timestamp rhythm of one frame every 50 milliseconds. The extraction process requires searching backward from the start frame 500 milliseconds before the kick trigger until the voice input channel produces effective parsing semantics or an end state. In the voice processing channel, the start frame is usually marked as the starting position of the user's voice-guided voice, such as the first sound frame of "open" in "open the tailgate", and the end frame is the corresponding position of the sentence's end sound frame. If the sentence is parsed into 5 frames, its start and end frame numbers may be T203 to T207, then the time period for recording the voice command is 08:37:11.203 to 08:37:11.4 53. Then, based on the feedback from the parsing engine status, the speech recognition activity status needs to be marked to determine whether it is in a high-frequency recognition segment within the task state. The effective parsing frames per second can be used as a reference. If the number of recognized frames exceeds 15 frames within 1 second, it is set as a high-frequency label, otherwise it is set as a regular active label. The current state of the recognition channel is further recorded. The background noise and speech recognition delay data in the same time period can be combined to assist in judging the activity state. For example, in the above sentence, the background noise recognition level is less than 30 decibels, the speech rate is 3.2 words per second, and the recognition delay is 150 milliseconds, all of which meet the regular activity requirements and can be marked as a standard task frame segment state. Finally, the start frame, end frame and activity state are encapsulated into a structured recognition record to obtain the speech recognition frame segment state information.

[0030] S302: Based on the speech recognition frame segment status information and the start and end frame numbers of the kick path, the speech and kick frame segments are time-aligned, the intersection segment between the start and end frames is extracted, and the number of overlapping frames in the intersection segment is calculated to obtain the number of overlapping frames of the input frame segment; The calculation formula for the number of overlapping frames in the crossover interval is: ; in, Represents the number of overlapping frames in the crossover interval, Represents each frame number within the crossover interval, Represents the start frame of the cross section, Represents the end frame number of the crossover segment, Representative The timestamp of the frame kick line, Representative Frame speech timestamp, Representative kick The path direction offset value of the frame, Representative voice The semantic parsing strength value of the frame, Represents the square value of the propagation stability of the path frame on the time axis, Represents the square value of the retention time of the speech recognition frame on the time axis.

[0031] Assumptions: The sampling period of each frame data is 40 milliseconds. When the sampling time is 120ms, ; Speech frame in When the sampling time is 100ms, ; In the standard path, the angle between the kicking direction and the center of gravity is used as the reference, and an angle less than 10° is considered as direction convergence. =0.8, =0.9; If the speed variation range is ±5cm / s, the stability value is 4. =16; The stable frame segment is 80ms, corresponding to a stability of 3.5. =12.25; Bring in frame Sample calculation when: ; The results show that the frame The intersection of the kicking trajectory and the speech recognition path shows a strong match, which can be included in the concurrent frame segment. The accumulation of values is the total number of overlapping frames in the intersection area, which is used to analyze the triggering conditions of the overlap of the interactive period, and then derive the value of the number of overlapping frames in the step result input frame segment.

[0032] S303: Based on the number of overlapping frames in the input frame segments, extract the frame segments as interaction evidence of temporal overlap between the voice input and the kicking path, set the frame segment status to concurrent input mark content, and obtain a multimodal control intervention state identifier; Extract the corresponding timestamp marks from the parsed frame segments of the voice command and the start and end frame segments of the kicking path, and establish a frame sequence comparison relationship under the unified timeline. The voice frame segment is usually represented by the frame number from T101 to T106, corresponding to a duration of 300 milliseconds, and the kicking path frame segment is T104 to T110, corresponding to a time coverage of 350 milliseconds. The two sequences have intersecting frame numbers between T104 and T106, forming a frame segment overlap interval. All frame numbers in the interval are extracted and included in the interaction basis judgment. If the number of frame segment overlaps reaches the set minimum overlap frame threshold, for example, 2 frames, the segment is considered to be a valid overlap interval, and then selected from the overlapping frame segments. The middle frame is taken as the representative frame segment, and the joint information of the frame in the speech analysis state and the kick path direction record is extracted. At the time point of this frame, the speech state must be in an active state (such as the activity label is "1"), and the kick path direction vector must be in a stable flow range, and the direction offset angle does not exceed 15 degrees. If the above conditions are met, it can be determined that the frame segment has the dual-channel synchronous interaction characteristics, and then the corresponding numbers T105 to T106 of the frame segment time window are recorded as concurrent interaction frame segments and set as concurrent input mark content. This content will be written into the multi-channel input state of the subsequent control judgment as the control source labeling condition to obtain the multimodal control intervention state identifier.

[0033] The specific steps of S4 are: S401: Based on the multimodal control intervention state identifier, extract the start and end frame numbers of the kick and voice channels, arrange the channel input sequences in frame segment order, and verify the channel source paths of the two sets of inputs based on the tag information to obtain the channel input sequence structure value; First, the frame segment number range bound to the voice and kick channels is extracted. For example, the start and end frame numbers extracted from the voice channel are T230 to T240, and the number range extracted from the kick path is T225 to T233. The two sets of frame segments are then imported into a unified time axis, and the frame segments are linearly sorted in chronological order in the time axis. During the sorting process, the inter-frame interval corresponding to the timestamp needs to be calculated frame by frame and a frame segment time series structure is generated. If the start frame time of the kick path frame segment on the time axis is earlier than the start time of the voice command parsing frame segment, The kick channel is set as the priority input path, and then combined with the marking information retained in the previous step, the information contains the source identification of each group of input channels, that is, the kick channel source number is marked as FT1, and the voice channel source number is VT1. After reading according to the rules, they are mapped and verified with the actual acquisition channels respectively. If the kick frame segment successfully matches the FT1 channel number and the voice frame segment successfully matches the VT1 number, it is considered that the channel source verification is consistent, and the source path is the valid calibration input path. The corresponding channel identifiers are further sequentially merged to obtain the channel input sequence structure value.

[0034] S402: Based on the channel input sequence structure value, extract the start frame number and the end frame number of the two channels, calculate the coverage span value between the first frame and the last frame of each channel, and extract the frame segment with complete coverage range as the candidate path based on the span value to obtain the channel frame segment coverage span; The calculation formula for the coverage span value between the first frame and the last frame of each channel is as follows: ; in, Representative Channel The coverage span value between the first frame and the last frame of Representative Channel The ending frame number, Representative Channel The starting frame number, Representative Channel The number of frame-segment pairs involved in the local structure weight analysis, Representative Channel No. The end frame number of the local frame segment, Representative Channel No. The starting frame number of a local frame segment, Representative Channel The average value of the start frame numbers of all local frame segments, Representative Channel The average value of the end frame number of the local frame segment, Representative Channel No. Weighting factors for local frame segments; Normalize the structural change indexes of all frame segments so that their sum is 1; The normalized value is the weight of the corresponding frame segment ; Assumptions: Frame segment 1: , ; Frame segment 2: , ; Frame segment 3: , ; Frame segment 1: ; Frame segment 2: ; Frame segment 3: ; ; ; 3. Formula calculation process: Frame segment 1: ; Frame segment 2: ; Frame segment 3: ; Frame segment 1: ; Frame segment 2: ; Frame segment 3: ; Frame segment 1: ; Frame segment 2: ; Frame segment 3: ; ; This result shows that the channel The frame segment coverage span value is 133.65, which indicates the degree of structural change from the start frame to the end frame. This value can be used to further analyze the dynamic change characteristics of the video content.

[0035] S403: Based on the channel frame segment coverage span, the channel corresponding to the coverage span value is set as the current master control path, and the input identification information corresponding to the channel is connected to the tailgate control interface to obtain the tailgate master control channel setting status; Call the start frame number and end frame number identified by the voice channel and kick channel in the previous step, perform span calculation on the frame segment range of the two groups of channels. If the kick channel frame segment is from T212 to T228, the span is 17 frames, and if the voice channel frame segment is from T215 to T222, the span is 8 frames. After recording, arrange them in order of span size and perform channel matching mapping. Set the channel with the larger span as the main control path, and then extract the input identification information attached to the channel. This information includes the channel type field, frame segment number, source tag, and status tag, etc. The setting format is such as "TY PE=FOOT, SEQ=T212-228, SRC=FT1", this input identifier is transmitted as the access content to the tailgate control channel entrance. In the channel entrance, the target path of the control port is registered according to the frame segment number of the input identifier and the identifier item set previously is overwritten. Then, combined with the current tailgate state hold flag and the vehicle unlocking condition, the path is identified as the current valid input channel path of the tailgate, thereby completing the switching update of the main control channel. This action process uses the identifier field as the execution unit, and connects the operation process between the vehicle state, input path and tailgate control in series. Get the tailgate master control channel setting status.

[0036] The specific steps of S5 are: S501: Based on the control source recorded in the tailgate master control channel setting state, the frame sequence number of the channel's current input control command is collected, and the static flag signal in the vehicle state is synchronously extracted and uniformly packaged into a path signal package to obtain a control input and state synchronization data set; First, extract the channel number and its bound control input frame segment number. The channel source information will mark whether the main control path is foot kick or voice input. If the channel is a foot kick path, the extracted frame segment number such as T302 to T312 means that the input signal frame segment coverage time is 500 milliseconds. Then, it is necessary to retrieve the static flag signal related to the tailgate at the current moment from the vehicle status acquisition. The flag signals involved mainly include ACC level status, door lock status, tailgate limit feedback status and tailgate unexecuted action identification. These signals are commonly configured as ACC low level, door lock unlocked, tailgate limiter closed and no last unfinished opening and closing action instruction is received in the tailgate waiting state. Each signal needs to extract the signal value corresponding to the input frame segment time period from the vehicle status interface according to the timestamp alignment method, and construct it into a time sequence state sequence, such as ACC state is L, door If the lock state is H, the limit state is H, and the control not executed is 1, they can be integrated into the state identifier "LHH-1". Then, the control input frame segment number and the state identifier are uniformly encapsulated to construct a path signal package. The fields of this signal package include: channel source, input frame segment number, current tailgate state value field, reception time mark, and signal validity label. In actual implementation, for example, if the kick channel input is from T302 to T312, the state identifier is "LHH-1", and the time is 08:46:53.285, then the path signal package format is "FT1|T302-312|LHH-1|08:46:53.285|1", where "FT1" represents the kick input channel and "1" is the signal validity identifier. After the integration is completed, the path signal package is saved as the current tailgate control context data input item, and the control input and state synchronization data set is obtained.

[0037] S502: Based on the control input and state synchronization data set, extract the tailgate stopper's closed feedback state and the displacement feedback sequence during the tailgate action execution process, and perform a frame-by-frame comparison between the displacement target value in the input control frame sequence and the feedback path to obtain control path correspondence information; First, read the control frame sequence number bound to the input instruction and the corresponding tailgate target opening and closing action state information. For example, the kick input frame segment is T312 to T320, and the corresponding instruction is "open". Then extract the closing feedback record value of the tailgate limiter in the current state. If the feedback value is closed, it indicates that the tailgate initial state is in the closed condition. Then enter the tailgate drive action execution stage, extract the tailgate displacement feedback sequence collected in this stage, and the displacement feedback records the opening and closing angle change value at a frequency of 0.2 seconds per frame. For example, the angle is 2 degrees at T313, 5 degrees at T314, and 10 degrees at T315. Expand in this way and record the angle change trend to reflect the physical movement process of the tailgate action. Next, compare the target displacement state of the input instruction with the feedback displacement in sequence. If the input target If the tailgate needs to open 15 degrees within 0.8 seconds and the current feedback sequence reaches 14 degrees at frame T316 and exceeds 15 degrees at frame T317, then this frame segment is considered to have a valid correspondence between the response path and the input path. The frame number range from T313 to T317 is used as the execution feedback interval for the input frame sequence. At the same time, the starting and ending state change amplitudes are identified in the feedback sequence. The linkage verification is combined with the level change point when the tailgate limiter closed signal changes to open. If the tailgate limiter signal changes from H to L at frame T317, the feedback path is confirmed to be time-consistent with the input path instruction. Finally, the frame number, state change point, action direction identifier, and other fields between the comparison sequence and the feedback sequence are integrated into the control path comparison record to obtain the control path correspondence information.

[0038] S503: Based on the control path correspondence information, the signal matching the control frame sequence structure in the feedback path is connected to the input port of the tailgate execution controller, the encapsulated input frame structure is identified, and the tailgate action process is scheduled after the control path is written to obtain the tailgate action management execution status identifier; First, confirm the intersection identifier of the matched control frame segment range and the previous instruction number in the feedback path. For example, the control path records frame segments T312 to T317 as the identified feedback segment. Here, the key frame number specifically used for tailgate opening and closing control in this segment should be extracted, and the segment data should be split into fields: frame number, action direction identifier, action trigger time, and action target angle. Then, the status of the current port to be received in the tailgate execution controller is read. When the controller receiving channel idle identifier is 1, the write operation of the control signal is started, and the number and angle value of each frame in the frame segment are mapped to the action displacement point in the tailgate motor drive requirement, encapsulated into a command data stream, and input into the controller in time sequence. For example, in frames T313 to T317, if the displacement value is 2 degrees, 5 degrees, 9 degrees, 14 degrees, and 17 degrees, then in order During writing, each piece of data is bound to a time tag, and an execution queue is built in the controller. The controller then identifies whether the input segment corresponds to the encapsulated input structure, that is, whether there is a channel tag with the field marked as "kick path" or "voice path". If the match is successful, the execution segment is classified as the current control task and connected to the tailgate action scheduling sequence. The motor drive direction is started according to the action direction field. If the field is "open", the driver sends a control current signal to the forward input line of the motor to control the motor to rotate in the opening direction. If the field is "off", it switches to the reverse input line. At the same time, the action process will trigger feedback recording, recording the motor displacement status and limit status level every 0.2 seconds. When the tailgate limit feedback changes from L to H, the execution signal is terminated, marking the end of the action, and the tailgate action management execution status identifier is obtained.

[0039] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A multimodal voice-controlled tailgate action management method based on vehicle state recognition, characterized in that: The following steps are involved: S1: Acquires the ACC level, door lock status, and tailgate closing signal, records the ACC interruption period, identifies the door lock unlocking status, and combines it with the tailgate closing feedback to process the intersection of the signal states and classify them into a stationary vehicle scenario to obtain the tailgate management activation status indicator; S2: Based on the tailgate management activation state flag, the foot trajectory is collected and the path start and end points and direction vectors are extracted. The trajectory convergence direction is matched with the offset difference and compared with the cursor area to obtain the kick control trigger instruction flag; S3: Based on the kick control trigger instruction identifier, extract the voice instruction parsing frame segment and mark the recognition status, retrieve the overlapping segment of the kick path and the voice frame sequence, locate the concurrent interaction state, and obtain the multimodal control intervention state identifier; S4: Based on the multimodal control intervention state identifier, extract the input frame sequence, select the channel of the first input structure as the main control path, inject it into the tailgate control interface, and obtain the tailgate main control channel setting state; S5: Based on the setting state of the tailgate master control channel, the command frame sequence and the static flag are input, and the corresponding displacement process path is fed back through the limit to obtain the tailgate action management execution state identifier.

2. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1 is characterized in that: The tailgate management activation status identifier includes the ACC interrupt hold time, level stability state parameters, door lock unlocking level type, tailgate closing feedback signal type, and signal combination classification flag; the kick control trigger instruction identifier includes the trajectory start and end time, Z-axis path change vector, convergence offset matching result, cursor overlap area number, and closed path annotation status; the multimodal control intervention status identifier includes the kick path number, voice frame segment number, voice active label, time overlap segment number, and interactive concurrency identification flag; the tailgate master control channel setting status includes the master control channel number, first segment complete structure mark, path continuity mark, input sequence positioning label, and control access channel identification code; the tailgate action management execution status identifier includes the input control frame segment number, static state matching flag, tailgate action feedback segment displacement sequence, feedback closure confirmation signal, and controller action execution status code.

3. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1 is characterized in that: The specific steps of S1 are: S101: Acquire the vehicle's current ACC level status, door lock status, and tailgate closed status, collect interruption points in the ACC level signal and mark the time period during which it is continuously maintained, read the unlocked state level signal type from the door lock control line, extract the closed feedback signal from the tailgate limit switch, and obtain a state signal acquisition record; S102: Based on the state signal collection record, extracting a signal combination of ACC hold interruption, door lock unlocking, and electric tailgate closing from the ACC level, door lock signal, and tailgate feedback, and comparing the state consistency of the signals within the same time period to obtain a static signal combination state; S103: Based on the static condition signal combination state, compare the static state trigger condition in the tailgate control scene, match the signal combination with the trigger condition, access the control entrance, set it to the tailgate control preparation state, and obtain the tailgate management activation state identifier.

4. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1 is characterized in that: The specific steps of S2 are: S201: Based on the tailgate management activation state identification period, Z-axis displacement values and corresponding timestamps within the continuous foot movement area are collected, the start and end time periods of the path are extracted, and the direction vectors between each frame are marked to obtain a foot path direction dataset; S202: Based on the foot path direction dataset, extract the starting direction and ending direction of the path, calculate the angle between the starting and ending points of the path, compare the angle with a convergence judgment angle reference value, identify the path shape that tends to converge in a single direction, and obtain a path convergence matching value; S203: Based on the path convergence matching value, the trajectory boundary coordinate points are extracted and compared with the boundary range of the projection area, and the interactive path of the trajectory forming a closed structure within the boundary range is identified to obtain the kick control trigger instruction identifier.

5. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 4 is characterized in that: The calculation formula of the angle between the starting and ending points of the path is specifically: ； in, Represents the path starting and ending point angle between the direction vectors of the path starting point and the end point, Represents the direction vector of the starting point of the path, Represents the direction vector of the path end point, Represents the sum of the displacement increments of the segments in the path in the x-axis direction, Represents the average value of the displacement increment of the segments in the path in the y-axis direction, Represents the standard deviation of the displacement increment in the x-axis direction of the path, Represents the standard deviation of the displacement increment in the y-axis direction of the path.

6. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1 is characterized in that: The specific steps of S3 are: S301: Based on the kick control trigger instruction identifier, extract the instruction parsing frame segment in the current speech recognition channel, record the start frame and end frame numbers of the speech input, and mark the active tag of the current recognition state to obtain speech recognition frame segment state information; S302: Based on the speech recognition frame segment state information, according to the start and end frame numbers of the kick path, the speech and kick frame segments are time-correlated, the intersection segment between the start and end frames is extracted, and the number of overlapping frames in the intersection segment is calculated to obtain the number of overlapping frames of the input frame segment; S303: Based on the number of overlapping frames of the input frame segments, extract the frame segments as the interaction basis for the overlap of the voice input and the kicking path in time sequence, set the frame segment status to the concurrent input mark content, and obtain the multimodal control intervention state identifier.

7. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 6 is characterized in that: The calculation formula for the number of overlapping frames in the crossover interval is specifically: ； in, Represents the number of overlapping frames in the crossover interval, Represents each frame number within the crossover interval, Represents the start frame of the cross section, Represents the end frame number of the crossover segment, Representative The timestamp of the frame kick line, Representative Frame speech timestamp, Representative kick The path direction offset value of the frame, Representative voice The semantic parsing strength value of the frame, Represents the square value of the propagation stability of the path frame on the time axis, Represents the square value of the retention time of the speech recognition frame on the time axis.

8. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1 is characterized in that: The specific steps of S4 are: S401: Based on the multimodal control intervention state identifier, extract the start and end frame numbers of the kick and voice channels, arrange the channel input sequences in frame segment order, and verify the channel source paths of the two groups of inputs in combination with the tag information to obtain the channel input sequence structure value; S402: Based on the channel input sequence structure value, extract the start frame number and the end frame number of the two channels, calculate the coverage span value between the first frame and the last frame of each channel, and extract the frame segment with complete coverage range as the candidate path according to the span value to obtain the channel frame segment coverage span; S403: Based on the channel frame segment coverage span, the channel corresponding to the coverage span value is set as the current main control path, and the input identification information corresponding to the channel is connected to the tailgate control interface to obtain the tailgate main control channel setting state.

9. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 8, characterized in that: The calculation formula for the coverage span value between the first frame and the last frame of each channel is specifically: ； in, Representative Channel The coverage span value between the first frame and the last frame of Representative Channel The ending frame number, Representative Channel The starting frame number, Representative Channel The number of frame-segment pairs involved in the local structure weight analysis, Representative Channel No. The end frame number of the local frame segment, Representative Channel No. The starting frame number of a local frame segment, Representative Channel The average value of the start frame numbers of all local frame segments, Representative Channel The average value of the end frame number of the local frame segment, Representative Channel No. The weighting factor of each local frame segment.

10. The multimodal voice-controlled tailgate action management method based on vehicle state recognition according to claim 1, characterized in that: The specific steps of S5 are: S501: Based on the control source recorded in the tailgate master control channel setting state, the frame sequence number of the channel's current input control command is collected, and the static flag signal in the vehicle state is synchronously extracted and uniformly packaged into a path signal package to obtain a control input and state synchronization data set; S502: Based on the control input and state synchronization data set, extract the closing feedback state of the tailgate stopper and the displacement feedback sequence during the tailgate action execution process, and perform frame-segment sequential comparison between the displacement target value in the input control frame sequence and the feedback path to obtain control path correspondence information; S503: Based on the control path correspondence information, the signal matching the control frame sequence structure in the feedback path is connected to the input port of the tailgate execution controller, the encapsulated input frame structure is identified, and the tailgate action process is scheduled after the control path is written to obtain the tailgate action management execution status identifier.

Images