An intelligent controller for an intelligent driving automobile

Abstract

The application relates to the technical field of intelligent controllers and discloses an intelligent controller of an intelligent driving automobile, which comprises a power module, a PDC module and an image processing module, wherein the power module is used for being connected with a vehicle-mounted power supply and supplying power for each module in the intelligent controller; the PDC module is used for being connected with an ultrasonic radar on the automobile, receiving and processing obstacle distance information; the image processing module is used for being connected with a visual camera on the automobile, receiving and preprocessing real-time image data; and an independent memory is used for pre-storing historical micro-control instruction data of the vehicle, which is collected by an original visual camera before replacement of components and verified. The closed-loop calibration mechanism composed of the independent memory, a calibration analysis module and the image processing module is constructed, adaptive alignment of the replaced visual camera or the intelligent controller is realized under the premise of not depending on external calibration equipment, and deviation between image data and control logic caused by replacement of components is effectively eliminated.

Description

Technical Field

[0001] This invention relates to the field of intelligent controller technology, and more particularly to an intelligent controller for an intelligent driving vehicle. Background Technology

[0002] Even after repairs, the vehicle continued to experience dangerous incidents such as sudden turns and sharp steering wheel movements, but multiple inspections failed to identify the root cause, and technicians were unable to guarantee subsequent safety, leaving the safety hazard as an unresolved "ghost."

[0003] Furthermore, in existing technologies, when replacing vision cameras or intelligent controllers during the maintenance of intelligent driving vehicles, problems such as image distortion, viewing angle deviation, or mismatch of control logic parameters between the new components and the original system can lead to imperceptible systematic deviations between perceived data and control commands. These deviations often cannot be effectively identified through conventional detection methods, but may cause sudden dangerous situations such as unexplained sharp turns or sudden steering wheel swerves in specific scenarios, leaving safety hazards lurking for a long time and causing great trouble for the safety of the repaired vehicle and the troubleshooting of technicians. Summary of the Invention

[0004] This invention provides an intelligent controller for intelligent driving vehicles to solve existing technical problems, thereby resolving the perception and control deviations caused by component differences after replacing the vision camera or intelligent controller in intelligent driving vehicles.

[0005] To solve the above-mentioned technical problems, according to one aspect of the present invention, more specifically, an intelligent controller for an intelligent driving vehicle, comprising:

[0006] The power module is used to connect to the vehicle power supply and power the various modules inside the intelligent controller;

[0007] The PDC module is used to connect to the ultrasonic radar on the vehicle to receive and process obstacle distance information.

[0008] The image processing module is used to connect to the vision camera on the vehicle to receive and preprocess real-time image data;

[0009] Independent memory is used to pre-store historical micro-control command data of the vehicle that was collected and verified by the original vision camera before the parts were replaced;

[0010] The calibration analysis module is connected to the image processing module and an external independent memory, respectively. It is used to compare and analyze the real-time image data transmitted by the image processing module with the micro-control command data made by the MCU micro-control module after the vehicle's vision camera or intelligent controller is replaced.

[0011] The image processing module is also used to adjust and compensate the input data of the currently connected visual camera according to the calibration parameters, and output the calibrated image data.

[0012] The MCU stability control module is connected to the calibration analysis module and the PDC module respectively, and is used for communication with the vehicle's body control module, hybrid power control unit and high voltage control unit;

[0013] The MCU stability control module is used to collect calibrated image data and obstacle distance information, generate micro-control command data after fusion processing, and send it to the body control module, hybrid power control unit and high voltage control unit.

[0014] Furthermore, the image processing module includes a digital signal processor, used to perform pixel remapping or affine transformation on the raw image data input from the vision camera according to the calibration parameters generated by the calibration analysis module, so as to eliminate image data deviation caused by component replacement.

[0015] Furthermore, the calibration analysis module constructs a calibration judgment model based on the deviation between historical microcontroller command data and the image data that generated the historical microcontroller command data, and performs the following steps:

[0016] S1. Input the acquired real-time micro-control command data and real-time image data into the calibration judgment model;

[0017] S2. Determine the existence of system deviation by checking whether the coefficients output by the calibration model exceed the preset values;

[0018] S3. If the coefficient output by the calibration judgment model does not exceed the preset value, the process ends; if the coefficient output by the calibration judgment model does not exceed the preset value, the calibration parameters for the corresponding gear are matched according to the coefficient output by the calibration judgment model.

[0019] S4. After feeding back the calibration parameters to the MUCH microcontroller module, repeat steps S1 to S4 until the coefficients output by the calibration judgment model do not exceed the preset value.

[0020] Furthermore, the deviation between the historical micro-control instruction data and the image data of the historical micro-control instruction data specifically includes the instruction reproduction deviation index and the image feature space offset index;

[0021] The instruction reproduction deviation index measures the degree of difference between the micro-control instructions generated by the current system in the same or similar scenarios and the historically verified micro-control instructions. It directly reflects the deviation of the control logic output after the controller or camera is replaced;

[0022] The image feature spatial offset index is used to measure the degree of geometric offset between the feature points of the image captured by the current vision camera and the feature points of the historical image in spatial coordinates, reflecting the perceptual differences caused by image distortion, viewing angle deviation or installation position changes.

[0023] Furthermore, the calibration judgment model is based on the deviation relationship between the historical microcontroller instruction data and the corresponding historical image data pre-stored in the independent memory. It is a mathematical mapping relationship model constructed by analyzing two quantitative indicators, the instruction reproduction deviation index and the image feature space offset index, and combining the multi-level calibration parameters used by the original controller before the replacement of the components, through the fitting of control variables and data.

[0024] This calibration judgment model is used to map the degree of deviation between the real-time acquired micro-control command data and image data to the corresponding calibration parameter output level, thereby realizing the quantitative judgment of system deviation and adaptive matching of calibration parameters.

[0025] Furthermore, the current image feature points are extracted from the image data, and the spatial coordinates of the current image feature points are compared with those of historical feature points in an independent memory.

[0026] Calculate the distortion correction coefficients used to correct image distortion or viewpoint deviation, and use these distortion correction coefficients as calibration parameters for the corresponding settings.

[0027] Furthermore, the PDC module is connected to the MCU stability control module via a CAN or LIN bus to transmit the processed obstacle distance information to the MCU stability control module in real time.

[0028] Furthermore, the MCU stability control module integrates a decision fusion algorithm, which, after receiving calibrated image data, combines the obstacle distance data from the PDC module to comprehensively judge the vehicle's surrounding environment and generate micro-control command data for adaptive cruise control, automatic emergency braking, or automatic parking functions.

[0029] Furthermore, the MCU stability control module interacts with the body control module in real time via CAN or LIN bus, interacts with the hybrid power control unit in real time via CAN or CAN FD bus, and interacts with the high-voltage control unit in real time via TPF bus.

[0030] Furthermore, the micro-control commands include drive torque commands, braking force commands, and lighting control commands.

[0031] The present invention provides an intelligent controller for a smart driving vehicle. Compared with the prior art, the advantages achieved by this method are as follows:

[0032] 1. This invention constructs a closed-loop calibration mechanism consisting of an independent memory, a calibration analysis module, and an image processing module. This enables adaptive alignment of the replaced vision camera or intelligent controller without relying on external calibration equipment. It effectively eliminates the deviation between image data and control logic caused by component replacement, fundamentally solving the problem of difficult-to-trace "ghost faults" such as vehicles making sudden turns or sharp steering wheel movements after maintenance. This significantly improves the reliability and safety of the system after maintenance.

[0033] 2. This invention introduces the instruction reproduction deviation index and the image feature space offset index as multi-dimensional quantitative evaluation indicators, transforming the originally difficult-to-capture system deviation into measurable and comparable mathematical parameters. This not only accurately identifies the specific sources of distortion at the sensing end and mismatch at the control logic end, but also provides clear data basis for the generation of calibration parameters, greatly improving the scientificity, interpretability and accuracy of the calibration process.

[0034] 3. This invention constructs a calibration judgment model based on the control variable method and data fitting, which realizes adaptive matching and hierarchical adjustment of calibration parameters, making the calibration process predictable and convergent. It can automatically approach the optimal calibration state within a finite number of iterations, effectively avoiding the inefficient operation of repeated trial and error based on experience in the traditional method, reducing the cost of manual calibration and debugging, and enhancing the operational consistency of the intelligent driving system throughout its entire life cycle. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the present invention;

[0036] Figure 2 This is a graph showing the relationship between the instruction reproduction deviation index and the assigned coefficient in this invention.

[0037] Figure 3 This is a graph showing the relationship between the image feature space offset index and the assigned coefficients in this invention.

[0038] Figure 4 This is a graph showing the relationship between the instruction reproduction deviation index and the image feature space offset index in this invention.

[0039] Figure 5 This is a flowchart of the present invention. Detailed Implementation

[0040] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] Example 1:

[0042] like Figure 1 , Figure 5As shown, according to one aspect of the present invention, an intelligent controller for an intelligent driving vehicle is provided, comprising: a power supply module for connecting to an on-board power supply and supplying power to various modules within the intelligent controller; a PDC module for connecting to an ultrasonic radar on the vehicle and receiving and processing obstacle distance information; an image processing module for connecting to a vision camera on the vehicle and receiving and preprocessing real-time image data; an independent memory for pre-storing historical micro-control command data collected and verified by the original vision camera before the vehicle's components were replaced; a calibration analysis module connected to the image processing module and the external independent memory, respectively, for receiving real-time image data transmitted by the image processing module and comparing and analyzing micro-control command data issued by the MCU micro-control module after the vehicle's vision camera or intelligent controller is replaced; the image processing module is also used to adjust and compensate the input data of the currently connected vision camera according to calibration parameters and output calibrated image data; and an MCU stability control module connected to the calibration analysis module and the PDC module, and for communicating with the vehicle's body control module, hybrid power control unit, and high-voltage control unit.

[0043] The MCU stability control module is used to collect calibrated image data and obstacle distance information, generate micro-control command data after fusion processing, and send it to the body control module, hybrid power control unit and high voltage control unit.

[0044] In this embodiment, the current image feature points are extracted from the image data, and the spatial coordinates of the current image feature points are compared with those of historical feature points in an independent memory. A distortion correction coefficient is calculated to correct image distortion or viewpoint deviation, and this distortion correction coefficient is used as the calibration parameter for the corresponding gear.

[0045] In this embodiment, the PDC module is connected to the MCU stability control module via a CAN or LIN bus, and is used to transmit the processed obstacle distance information to the MCU stability control module in real time. The MCU stability control module integrates a decision fusion algorithm, which, after receiving calibrated image data, combines it with the obstacle distance data from the PDC module to comprehensively judge the vehicle's surrounding environment and generate micro-control command data for adaptive cruise control, automatic emergency braking, or automatic parking functions. The MCU stability control module interacts with the body control module in real time via a CAN or LIN bus, with the hybrid power control unit via a CAN or CAN FD bus, and with the high-voltage control unit via a TPF bus.

[0046] In this embodiment, the micro-control commands include drive torque commands, braking force commands, and lighting control commands.

[0047] The core innovation of this embodiment lies in constructing a complete closed-loop calibration mechanism. It utilizes a dedicated memory to pre-store verified historical microcontroller command data and image feature points before component replacement, serving as a calibration benchmark. After replacing the vision camera or intelligent controller, the calibration analysis module compares real-time image data with the microcontroller command data generated by the MCU stabilization module to identify perception and control deviations caused by component replacement, and drives the image processing module to adjust and compensate the input data from the current camera. This mechanism achieves adaptive alignment between the controller and the camera without relying on external calibration equipment, ensuring the original consistency between image data and control decisions.

[0048] Its advantages lie in effectively solving the "ghost failure" problem caused by camera or controller replacement after vehicle repair, such as unexpected sharp turns or sudden steering wheel movements. Through an automated and standardized calibration process, it avoids the predicament of technicians repeatedly testing without being able to pinpoint the root cause, significantly improving the reliability and safety of the system after repair, and ensuring the continued stable operation of intelligent driving functions after component replacement.

[0049] Example 2:

[0050] like Figure 1 , Figure 5 As shown, according to one aspect of the present invention, an intelligent controller for an intelligent driving vehicle is provided, wherein the image processing module includes a digital signal processor for performing pixel remapping or affine transformation on the raw image data input from the vision camera according to calibration parameters generated by the calibration analysis module, so as to eliminate image data deviation caused by component replacement.

[0051] The calibration analysis module constructs a calibration judgment model based on the deviation between historical microcontroller command data and the image data that generated the historical microcontroller command data, and performs the following steps:

[0052] S1. Input the acquired real-time micro-control command data and real-time image data into the calibration judgment model;

[0053] S2. Determine the existence of system deviation by checking whether the coefficients output by the calibration model exceed the preset values;

[0054] S3. If the coefficient output by the calibration judgment model does not exceed the preset value, the process ends; if the coefficient output by the calibration judgment model does not exceed the preset value, the calibration parameters for the corresponding gear are matched according to the coefficient output by the calibration judgment model.

[0055] S4. After feeding back the calibration parameters to the MUCH microcontroller module, repeat steps S1 to S4 until the coefficients output by the calibration judgment model do not exceed the preset value.

[0056] The deviation between historical microcontroller command data and the image data of that historical microcontroller command data specifically includes the command reproduction deviation index and the image feature space offset index;

[0057] Instruction Reproduction Deviation Index This is used to measure the degree of difference between the microcontroller commands generated by the current system in the same or similar scenarios and historically verified microcontroller commands. It directly reflects the deviation of the control logic output after the controller or camera is replaced, specifically:

[0058] ;

[0059] In the above formula, This represents the microcontroller instruction vector generated by the current system in the j-th key scenario; This represents the historical microcontroller instruction vector corresponding to the scenario in the independent memory; Indicates the number of key scenarios involved in the comparison; Represents the Euclidean distance, and calculates the overall difference between command vectors; This indicates the permissible threshold for non-sensory deviation; This represents the sensitivity adjustment coefficient.

[0060] Image feature space offset index This is used to measure the geometric offset of feature points in the image acquired by the current vision camera from feature points in historical images in terms of spatial coordinates. It reflects the perceptual differences caused by image distortion, viewing angle deviation, or changes in installation position. Specifically:

[0061] ;

[0062] In the above formula, This represents the position of the i-th current image feature point in the normalized spatial coordinate system; This indicates the location of the corresponding historical image feature point in the independent memory; Indicates the number of matched feature points; This indicates the maximum allowed offset distance (e.g., half the length of the image diagonal).

[0063] This embodiment further quantifies the deviation identification process by introducing two multi-dimensional evaluation indicators: the "command reproduction deviation index" and the "image feature space offset index," which measure system deviation from the perspectives of control logic consistency and perceptual data geometric consistency, respectively. By comparing the spatial coordinates of image feature points and calculating the difference between historical microcontroller command vectors and the current command vector, a refined identification of the source of deviation is achieved, providing a mathematical basis for the generation of subsequent calibration parameters.

[0064] Its advantage lies in transforming abstract system deviations into quantifiable and comparable indicators, making calibration judgments interpretable and reproducible. The parallel evaluation of dual indicators can distinguish between deviations caused by distortion at the sensing end and mismatches at the control logic end, thereby guiding the generation of calibration parameters more specifically, improving the scientific rigor and accuracy of the calibration process, and laying a data foundation for the construction of subsequent calibration models.

[0065] Example 3:

[0066] like Figures 1-5 As shown, according to one aspect of the present invention, the specific steps for constructing a calibration judgment model in an intelligent controller of an intelligent driving vehicle are as follows:

[0067] Step S1: Obtain the historical microcontroller instruction data in the independent memory, the instruction reproduction deviation index and image feature space offset index between the image data that generated the historical microcontroller instruction data, and the calibration parameters given by the original intelligent controller when the vision camera or intelligent controller was not replaced.

[0068] The calibration parameters given by the original controller are sorted according to their calibration degree, and the difference between two adjacent calibration parameters is between 0.1% and 0.4%. For example: The tangential distortion coefficient is always set to 0.01 (the tangential distortion coefficient ranges from -0.01 to 0.01). The tangential distortion coefficient is obtained by using it as a fixed control variable in the calibration model, selecting a typical value within the calibration range (e.g., 0.01), and based on historical image data stored in an independent memory, analyzing the spatial offset patterns of image feature points under different tangential distortion coefficients, and combining the control variable method, fitting the mapping relationship between the tangential distortion coefficient and the image feature spatial offset index, thus determining the value of the tangential distortion coefficient in the model and its influence weight on the calibration parameters). Then, radial distortion coefficients are sampled according to -0.99, -0.98, ..., 0, ..., 0.98, 0.99, where the radial distortion coefficient ranges from -0.5 to 0.5. The collected samples are sorted by absolute value and assigned coefficients according to the sort order. Then:

[0069] Table 1. Obtaining the absolute value of the radial distortion coefficient by controlling the tangential distortion coefficient.

[0070]

[0071] For example, based on the radial distortion coefficients of the positive calibration in Table 1 above, the instruction reproduction deviation index and image feature space offset index for the corresponding sample number can be obtained. Furthermore, based on the relationship between the instruction reproduction deviation index, the image feature space offset index, and the assigned coefficients, the relationship between the radial distortion coefficients of the positive calibration (negative calibration) and the instruction reproduction deviation index and image feature space offset index can be constructed when the tangential distortion coefficient is 0.01.

[0072] Step S2: Further, the image feature space offset index is controlled by the controlled variable method. In order to determine the instruction reproduction deviation index The relationship between the assigned coefficients (positive calibration);

[0073] If we further filter from the above samples, we can filter for image feature space offset index. Samples with all values ​​of 0.34 (e.g., the most common value among all samples is 0.34) are assigned coefficients as follows: Then the instruction reproduction deviation index With the assigned coefficient The mathematical expression between them is:

[0074] ;

[0075] In the above formula, , , Used for control and The coefficients tend to approximate the given values. Furthermore, by analyzing the aforementioned samples, the coefficients in the formula are determined. , , The value of . For example, by Figure 2 Able to determine , , When, in the formula and Approaching approximation.

[0076] Step S3: Further, the instruction reproduction deviation index is controlled again using the controlled variable method. This allows us to determine the image feature space offset index. The relationship between the assigned coefficients (positive calibration);

[0077] The sample was further filtered again, and the filtering command reproduced the deviation index. The samples are all 0.34, and the corresponding assigned coefficients are denoted as... Then the image feature space offset index With the assigned coefficient The mathematical expression between them is:

[0078] ;

[0079] In the above formula, , , Used for control and The coefficients tend to approximate the given values. Furthermore, by analyzing the aforementioned samples, the coefficients in the formula are determined. , , The value of . For example, by Figure 3 Able to determine , , When, in the formula and Approaching approximation.

[0080] Step S4: Using the relationships from the above steps, a public data quality assessment model corresponding to the above assessment rules is fitted and generated. Then, the sample from step S1 is further filtered, and the filtered data must include the coefficients assigned in steps S2 and S3. Furthermore, the assigned coefficients in this filtering process must be consistent, and they are denoted as follows: .pass Deviation index from instruction reproduction Image feature space offset index To fit the relationship between the two, we can obtain a mathematical expression for the calibration judgment model of positive calibration under the condition of tangential distortion coefficient of 0.01. Then we have:

[0081] ;

[0082] Similarly, in the above formula , Used for control and A constant that approaches an approximation. And by analyzing the above samples, the value in the formula is further determined. , The value of . For example, by Figure 5 Able to determine , When, in the formula and Approaching approximation.

[0083] So, according to Figures 2-4 The mathematical expression for the calibration judgment model generated by the mid-parameter fitting for positive calibration is:

[0084] ;

[0085] Then, the acquired real-time micro-control command data and real-time image data are input into the calibration judgment model; and the existence of system deviation is judged based on whether the coefficients output by the calibration judgment model exceed the preset value. At this point, we have:

[0086] Table 2 Model output coefficients and corresponding calibration parameters

[0087]

[0088] This embodiment constructs a calibration judgment model based on the control variable method and data fitting. It establishes functional relationships between the instruction reproduction deviation index and the image feature space offset index under different distortion coefficients in historical samples, establishing these relationships step-by-step. Finally, these are merged to form a unified calibration judgment model. This model can automatically match the corresponding calibration parameters based on the two deviation indices currently input by the system, and gradually approach the optimal calibration state through an iterative feedback mechanism.

[0089] Its advantages lie in achieving adaptive matching and hierarchical adjustment of calibration parameters, avoiding the inefficient process of repeated trial and error based on experience in traditional methods. By quantifying the calibration parameters in segments and establishing a mathematical model, the calibration process becomes predictable and convergent, enabling the effective elimination of system deviations within a finite number of iterations. This method significantly improves the adaptability of the intelligent controller after component replacement, reduces the cost of manual calibration and debugging, and enhances the stability and consistency of the intelligent driving system throughout its entire lifecycle.

[0090] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An intelligent controller for an intelligent driving vehicle, characterized in that, include: The power module is used to connect to the vehicle power supply and power the various modules inside the intelligent controller; The PDC module is used to connect to the ultrasonic radar on the vehicle to receive and process obstacle distance information. The image processing module is used to connect to the vision camera on the vehicle to receive and preprocess real-time image data; Independent memory is used to pre-store historical micro-control command data of the vehicle that was collected and verified by the original vision camera before the parts were replaced; The calibration analysis module, connected to both the image processing module and an external independent memory, is used to compare and analyze real-time image data transmitted from the image processing module with micro-control command data issued by the MCU micro-control module after the vehicle's vision camera or intelligent controller is replaced. The calibration analysis module constructs a calibration judgment model based on the deviation between historical micro-control command data and the image data that generated the historical micro-control command data, and performs the following steps: S1. Input the deviation between the acquired real-time micro-control command data and the real-time image data into the calibration judgment model; S2. Determine the existence of system deviation by checking whether the coefficients output by the calibration model exceed the preset values; S3. If the coefficient output by the calibration judgment model does not exceed the preset value, the process ends; if the coefficient output by the calibration judgment model exceeds the preset value, the calibration parameters for the corresponding gear are matched according to the coefficient output by the calibration judgment model. S4. After feeding the calibration parameters back to the MUCH microcontroller module, repeat steps S1 to S4 until the coefficients output by the calibration judgment model do not exceed the preset value. The image processing module is also used to adjust and compensate the input data of the currently connected visual camera according to the calibration parameters, and output the calibrated image data. The MCU stability control module is connected to the calibration analysis module and the PDC module respectively, and is used for communication with the vehicle's body control module, hybrid power control unit and high voltage control unit; The MCU stability control module is used to collect calibrated image data and obstacle distance information, generate micro-control command data after fusion processing, and send it to the body control module, hybrid power control unit and high voltage control unit.

2. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: The image processing module includes a digital signal processor, which performs pixel remapping or affine transformation on the raw image data input from the vision camera based on the calibration parameters generated by the calibration analysis module, so as to eliminate image data deviation caused by component replacement.

3. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: The deviation between the historical microcontroller command data and the image data of the historical microcontroller command data specifically includes the command reproduction deviation index and the image feature space offset index; The instruction reproduction deviation index is used to measure the degree of difference between the micro-control instructions generated by the current system in the same or similar scenarios and the historically verified micro-control instructions. The image feature spatial offset index is used to measure the degree of geometric offset between the feature points of the image captured by the current vision camera and the feature points of the historical image in spatial coordinates, reflecting the perceptual differences caused by image distortion, viewing angle deviation or installation position changes.

4. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: The calibration judgment model is based on the deviation relationship between the historical microcontroller instruction data and the corresponding historical image data pre-stored in the independent memory. It is a mathematical mapping relationship model constructed by analyzing two quantitative indicators, the instruction reproduction deviation index and the image feature space offset index, and combining the multi-level calibration parameters used by the original controller before the replacement of the components, through the fitting of control variables and data.

5. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: Extract the current image feature points from the image data and compare the spatial coordinates of the current image feature points with those of historical feature points in an independent memory. Calculate the distortion correction coefficients used to correct image distortion or viewpoint deviation, and use these distortion correction coefficients as calibration parameters for the corresponding settings.

6. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: The PDC module is connected to the MCU stability control module via a CAN or LIN bus, and is used to transmit the processed obstacle distance information to the MCU stability control module in real time.

7. The intelligent controller for an intelligent driving vehicle according to claim 1, characterized in that: The MCU stability control module integrates a decision fusion algorithm, which, after receiving calibrated image data, combines obstacle distance data from the PDC module to comprehensively judge the vehicle's surrounding environment and generate micro-control command data for adaptive cruise control, automatic emergency braking, or automatic parking functions.

8. The intelligent controller for an intelligent driving vehicle according to claim 7, characterized in that: The MCU stability control module interacts with the body control module in real time via CAN or LIN bus, interacts with the hybrid power control unit in real time via CAN or CAN FD bus, and interacts with the high-voltage control unit in real time via TPF bus.

9. The intelligent controller for an intelligent driving vehicle according to claim 7, characterized in that: The micro-control commands include drive torque commands, braking force commands, and lighting control commands.

Images