Preparation method of intelligent vehicle circuit board based on vision

Abstract

The invention discloses a preparation method of an intelligent vehicle circuit board based on vision, and relates to the technical field of intelligent driving. The method comprises the following steps that 1, a base material is pretreated, specifically, a high-Tg halogen-free FR-4 base material with the thickness of 1.6 mm and the surface gold immersion layer thickness of 10 micrometers is selected, surface treatment is conducted through a plasma cleaning technology, the cleaning power is 300 W, the treatment time is 15 s, and the water contact angle of the surface of the base material is smaller than or equal to 30 degrees; step 2, manufacturing a circuit pattern, adopting an LDI laser direct imaging technology to carry out circuit exposure, enabling the line width and the line distance to be 0.1 mm / 0.1 mm and the positioning precision to be + / -0.01 mm, and adopting an acidic copper chloride etching solution to carry out etching at 45 + / -2 DEG C and the etching speed to be 1.2 mu m / min; according to the invention, through integrated wiring design and special interface integration, visual perception, data processing and vehicle control are highly integrated on a single board, and a special visual processing unit is combined, so that the multi-channel visual data processing delay can be reduced to be less than 50ms, such as 42ms in the embodiment, and the real-time response capability of an intelligent driving system is greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving, in particular to a preparation method of an intelligent vehicle circuit board based on vision. BACKGROUND

[0002] With the evolution of intelligent driving technology to a higher level, the visual recognition system has become the core module of environmental perception. The vehicle visual recognition intelligent driving control circuit board needs to meet three core requirements: first, high-speed processing of multi-channel high-definition visual data (such as 20GB+ data generated by an 8 million pixel camera per second); second, maintaining stability in a vehicle environment of extreme temperature from -40℃ to 125℃ and wide frequency vibration from 10Hz to 2000Hz; third, having functional safety redundancy and meeting the ASIL-D level safety requirements of ISO 26262 standard.

[0003] The existing vehicle visual recognition intelligent driving control circuit board has high data processing delay and insufficient real-time performance: the traditional circuit board uses a general processor architecture and lacks hardware acceleration units, and the processing delay of multi-camera synchronous data often exceeds 100ms, which cannot meet the millisecond-level decision-making requirements in emergency working conditions; the compatibility of vision and control is poor and the integration is low: the visual recognition module and the driving control module are often independent circuit boards, and data is transmitted through external buses, which causes signal attenuation and interface protocol mismatch, resulting in control command synchronization deviation; using ordinary FR-4 substrate and conventional wiring design, the substrate is prone to deformation and copper layer peeling (peeling strength less than 1.2N / mm) in high temperature environment, and the core temperature of high-performance chips is prone to exceed 130℃ when working, causing performance downshift.

[0004] Therefore, a preparation method of an intelligent vehicle circuit board based on vision is proposed. SUMMARY

[0005] The purpose of the present application is to solve the problems raised in the background art, and the present application provides a preparation method of an intelligent vehicle circuit board based on vision.

[0006] In order to achieve the above purpose, the present application specifically adopts the following technical scheme: A preparation method of an intelligent vehicle circuit board based on vision, comprising the following steps: Step 1: substrate pretreatment, high Tg halogen-free FR-4 substrate is selected, thickness 1.6mm, surface gold layer thickness 10μm, surface treatment is carried out by using plasma cleaning process, cleaning power 300W, treatment time 15s, so that the water contact angle of the substrate surface is ≤30°; Step two: circuit pattern making, using LDI laser direct imaging technology for circuit exposure, line width and distance is 0.1mm / 0.1mm, positioning accuracy ±0.01mm, using acidic copper chloride etching solution at 45±2℃ for etching, etching speed 1.2μm / min, and through AOI detection to ensure that the circuit has no defects; Step three: hole processing and metallization, using numerical control drilling machine to process micro-holes with minimum aperture of 0.2mm, chemical copper plating and electroplating copper treatment, hole wall copper thickness ≥25μm, key signal hole adopts blind buried hole design; Step four: component welding and assembly, using SMT patch process to weld surface components, reflow soldering temperature curve includes preheating zone, heating zone, peak zone and cooling zone, core chip welding is followed by bottom filling, using vehicle-grade epoxy resin to cure at 150℃ for 60min; Step five: strengthening treatment and test, coating 20-30μm thick three-proofing paint on the non-welding area of the circuit board, and then sequentially performing electrical performance test, environmental stress test and function verification test.

[0007] Further, the reflow soldering temperature curve in step four is: preheating zone: temperature is 150-180℃, time is 60s; heating zone: temperature is 180-220℃, time is 30s; peak zone: temperature is 245±5℃, time is 10s; cooling zone: temperature is less than 100℃, time is 60s.

[0008] Further, the electrical performance test in step five uses flying probe test with coverage of 100%, the environmental stress test includes high-low temperature cycle for 50 times and vibration test for 20 hours, and the function verification test includes at least 1000 times of target recognition and control response test by connecting to the real vehicle vision system.

[0009] Further, the method further comprises wiring design and optimization before step two, specifically including: Defining software and hardware interface protocol based on intelligent vehicle sensing, decision-making and control function, inputting to the comprehensive wiring module for verification; Defining signal layer connection relationship according to the wiring information of the double-sided circuit board, optimizing the packaging stacking mode, and generating a single-layer board; Through the layout editing module, the double-layer board interactive design is generated, and the final wiring information is generated.

[0010] Further, in the wiring design, the interface protocol includes at least one of CAN FD, Ethernet AVB, GMSL2, and the signal transmission bit error rate is ensured to be ≤ .

[0011] Further, the method further comprises an anti-short circuit level test on the circuit board, specifically comprising: Attaching metal conductive particles with different diameters on the surface of the copper layer as short circuit points; By simulating the particle sliding process, the surface state of the copper layer is characterized by SEM to determine the anti-short circuit level of the circuit board.

[0012] Further, the method further comprises an anti-vibration level test on the circuit board, specifically comprising: Building vibration experiment parameters, and characterizing the stress distribution of the interface between the copper layer and the medium layer under different vibration frequencies by SEM; Combined with the conductivity test, the anti-shear level of the circuit board is determined.

[0013] Further, the conductivity test specifically comprises: Divide the circuit board into equilateral triangle regions with a side length of 5-10 cm; Use a multimeter to measure the conductivity between the triangle vertices and the center point; According to the conductivity data distribution, the anti-shear level of the circuit board is determined.

[0014] Further, the method further comprises an impedance consistency test on the circuit board, specifically comprising: Measuring the conductivity of each signal layer; Calculating the average impedance, impedance non-uniformity and impedance variation coefficient; Detecting via diameter and precision, copper layer interlayer coplanarity and medium layer flatness.

[0015] Further, the circuit board comprises a core control unit, a visual signal interface module, a multi-source fusion and control output unit, an environmental self-adaptive protection unit and an expandable storage and upgrading unit; the core control unit adopts a dual-core architecture of main computing power SoC and safety MCU, the main SoC is a vehicle-grade NVIDIA Jetson AGX Orin chip, and the safety MCU is an Infineon AURIX TC49x series chip; the visual signal interface module integrates 4 GMSL2 interfaces and 2 Ethernet AVB interfaces, and is provided with a signal equalizer and an EMC shielding circuit.

[0016] The beneficial effects of the present application are as follows: Through integrated wiring design and special interface integration, visual perception, data processing and vehicle control are highly integrated on a single board, and combined with a special visual processing unit, the multi-channel visual data processing delay can be reduced to within 50 ms, such as 42 ms in the embodiment, greatly improving the real-time response capability of the intelligent driving system; By adopting a series of automotive-grade processes such as high-Tg substrate, plasma cleaning, underfilling, three-protection coating, etc., the long-term stable operation of the circuit board under extreme temperature, mechanical vibration and harsh chemical environment is ensured. The copper layer peeling strength can reach 1.8 N / mm or more, which is much higher than the common level in the industry. The failure switching mechanism is verified through a strict test process, which meets the ASIL-D level functional safety requirements, the switching time is less than 10 ms, has extensible hardware interface and OTA upgrade capability, supports subsequent iteration of algorithms and sensors, avoids overall hardware replacement, and significantly reduces the upgrade cost of the whole vehicle life cycle. By introducing precise control and detection means such as LDI, AOI, plasma cleaning, etc. throughout the process, the preparation process is highly controllable and standardized, and the production defect rate of the circuit board can be stably controlled below 0.5%, meeting the quality requirements of large-scale mass production. BRIEF DESCRIPTION OF DRAWINGS

[0017] Fig. 1 is a process flow chart of the preparation method of the present application; Fig. 2 is a flow chart of the wiring design and optimization steps of the present application; Fig. 3 is a flow chart of the comprehensive test steps of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0020] It should be noted that: similar numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", etc. are only used for differentiation, and cannot be understood as indicating or implying relative importance.

[0021] In the description of the embodiments of the present application, it should be noted that the terms "inner", "outer", "upper", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the present application is usually placed, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0022] As shown in Figs. 1 to 3 A visual-based intelligent vehicle circuit board preparation method, comprising the following steps: Step one: substrate pretreatment, select high Tg halogen-free FR-4 substrate, thickness 1.6mm, surface gold layer thickness 10μm, adopt plasma cleaning process for surface treatment, cleaning power 300W, treatment time 15s, make substrate surface water contact angle≤30°; Step two: circuit pattern making, adopt LDI laser direct imaging technology for circuit exposure, line width line distance 0.1mm / 0.1mm, positioning accuracy ±0.01mm, adopt acidic copper chloride etching solution for etching at 45±2℃, etching speed 1.2μm / min, and ensure that the circuit has no defects through AOI detection; Step three: hole processing and metallization, adopt numerical control drilling machine to process micro-holes with minimum aperture 0.2mm, carry out chemical copper plating and electroplating copper processing, hole wall copper thickness≥25μm, key signal holes adopt blind buried hole design; Step four: component soldering and assembly, adopt SMT patch process to solder surface components, reflow soldering temperature curve includes preheating zone, heating zone, peak zone and cooling zone, after core chip soldering, carry out bottom filling, use vehicle-grade epoxy resin to solidify at 150℃ for 60min; The reflow soldering temperature curve in step four is: preheating zone: temperature 150-180℃, time 60s; heating zone: temperature 180-220℃, time 30s; peak zone: temperature 245±5℃, time 10s; cooling zone: temperature less than 100℃, time 60s.

[0023] Step five: strengthening treatment and test, coat 20-30μm thick three-proofing paint on the non-soldering area of the circuit board, after solidification, carry out electrical performance test, environmental stress test and function verification test in sequence; The electrical performance test in step five adopts flying probe test, coverage rate 100%, the environmental stress test includes high-low temperature cycle 50 times and vibration test 20 hours, the function verification test includes at least 1000 times target recognition and control response test by connecting to the real vehicle vision system.

[0024] The circuit board comprises a core control unit, a visual signal interface module, a multi-source fusion and control output unit, an environment self-adaptive guarantee unit and an extensible storage and upgrading unit; the core control unit adopts a dual-core architecture of a main computing power SoC and a safety MCU, the main SoC is a vehicle-grade NVIDIA Jetson AGX Orin chip, and the safety MCU is an Infineon AURIX TC49x series chip; the visual signal interface module integrates 4 GMSL2 interfaces and 2 Ethernet AVB interfaces, and is internally provided with a signal equalizer and an EMC shielding circuit.

[0025] In actual application, the method further comprises, before step two, wiring design and optimization, specifically comprising: Based on the intelligent vehicle sensing, decision-making and control functions, define the software and hardware interface protocol, input to the comprehensive wiring module for verification; According to the wiring information of the double-sided circuit board, define the connection relationship of the signal layer, optimize the packaging stacking mode, and generate a single-layer version; Through the layout editing module, interactive design of double-layer version is carried out, and the final wiring information is generated.

[0026] In the wiring design, the interface protocol includes at least one of CAN FD, Ethernet AVB and GMSL2, and the signal transmission error rate is ensured to be less than 10-12 through the signal equalizer and the EMC shielding circuit. ; The method further comprises anti-short circuit level testing of the circuit board, specifically comprising: Different diameter metal conductive particles are attached to the surface of the copper layer as short circuit points; Through simulation of the particle sliding process, the copper layer surface state is characterized by SEM, and the anti-short circuit level of the circuit board is determined.

[0027] The method further comprises anti-vibration level testing of the circuit board, specifically comprising: Vibration experiment parameters are constructed, and the stress distribution of the copper layer and the medium layer interface under different vibration frequencies is characterized by SEM; Combined with the conductivity test, the anti-shear level of the circuit board is determined.

[0028] The conductivity test specifically comprises: The circuit board is divided into equilateral triangle regions with a side length of 5-10 cm; The conductivity between the triangle vertices and the center point is measured using a multimeter.

[0029] The method further comprises impedance consistency detection of the circuit board, specifically comprising: Measure the conductivity of each signal layer; Calculate the average impedance, impedance non-uniformity and impedance variation coefficient; Detecting via diameter and precision, copper layer interlayer coplanarity and dielectric layer flatness.

[0030] Embodiments of the present application are not limited thereto; Example One; Preparation of L3 Level Intelligent Driving Visual Control Board 1. Hardware architecture design The circuit board comprises: Core control unit: NVIDIA Jetson AGX Orin (vehicle level) is used as the main computing SoC, and Infineon AURIX TC497 is used as the safety MCU.

[0031] Visual signal interface module: integrated 4-way GMSL2 interface based on MAXIM MAX96712 chip and 2-way Ethernet AVB interface, with built-in EMC shielding circuit.

[0032] Environment adaptive protection unit: integrated temperature sensor and intelligent heat dissipation control module.

[0033] 2. Preparation process Substrate pretreatment: halogen-free FR-4 substrate with Tg=180℃ is selected, thickness is 1.6mm, and gold deposition is 10μm. Plasma cleaning is carried out at 300W power for 15 seconds, and the water contact angle is measured to be 28°.

[0034] Circuit pattern making: LDI exposure is used, and the line width / line spacing is 0.1mm / 0.1mm. Acidic copper chloride etching solution is used for etching at 46℃, and AOI detection is error-free.

[0035] Hole metallization: via holes with a diameter of 0.25mm are drilled, and chemical copper deposition is followed by electroplating, and the copper thickness of the hole wall is measured to be 27μm.

[0036] Soldering and assembly: after SMT patching, soldering is carried out according to the predetermined reflow soldering curve. Bottom filling is carried out for Jetson AGX Orin and TC497 chips, LOCTITE 3536 epoxy resin is used, and curing is carried out at 150℃ for 60 minutes.

[0037] Strengthening and testing: selectively coat 25μm thick three-proof paint, and cure at 80℃. The flying probe test passes at a rate of 100%. After 50 cycles of-40℃~125℃ and 20 hours of vibration test, the circuit board functions normally. When connected to the real vehicle system, the processing delay is 42ms, and the fault switching time is 8ms.

[0038] 3. Test The finished circuit board is subjected to short circuit resistance experiment, metal conductive particles are applied in different copper layer regions, and SEM images are observed to determine that the short circuit resistance grade is L5. Through vibration table experiment and conductivity distribution test, the vibration resistance grade and shear resistance grade meet the vehicle requirement.

[0039] 4. Conclusion The circuit board prepared by the method has excellent real-time performance, reliability, safety and producibility, and is suitable for large-scale vehicle application.

[0040] In summary: through integrated wiring design and special interface integration, visual perception, data processing and vehicle control are highly integrated on a single board, and combined with a special visual processing unit, multi-channel visual data processing delay can be reduced to within 50ms, such as 42ms in the embodiment, greatly improving the real-time response capability of the intelligent driving system; by using a series of vehicle-grade processes such as high-Tg substrate, plasma cleaning, underfill, and three-proofing coating, the long-term stable operation of the circuit board in extreme temperature, mechanical vibration and harsh chemical environment is ensured. The copper layer peeling strength can reach 1.8N / mm or more, which is much higher than the common level in the industry. The failure switching mechanism is verified through a strict test process, which meets the ASIL-D level functional safety requirements, and the switching time can be less than 10ms. It has scalable hardware interfaces and OTA upgrade capability, supports subsequent iteration of algorithms and sensors, avoids overall hardware replacement, and significantly reduces the upgrade cost of the whole vehicle life cycle. Through the introduction of precise control and detection means such as LDI, AOI, and plasma cleaning throughout the process, the production process is highly controllable and standardized, and the production of the circuit board can be controlled to below 0.5%, meeting the quality requirements of large-scale production and vehicle application.

[0041] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection required by the present application is defined by the appended claims and their equivalents.

Claims

1. A method for fabricating a vision-based intelligent vehicle circuit board, characterized in that, Includes the following steps: Step 1: Substrate pretreatment. High Tg halogen-free FR-4 substrate with a thickness of 1.6mm and a surface gold plating layer thickness of 10μm is selected. Surface treatment is carried out using plasma cleaning process with a cleaning power of 300W and a treatment time of 15s, so that the water contact angle of the substrate surface is ≤30°. Step 2: Circuit pattern fabrication. LDI laser direct imaging technology is used for circuit exposure, with line width and spacing of 0.1mm / 0.1mm and positioning accuracy of ±0.01mm. Acidic copper chloride etching solution is used for etching at 45±2℃ with an etching speed of 1.2μm / min. AOI inspection is used to ensure that the circuit is defect-free. Step 3: Hole machining and metallization. Microholes with a minimum diameter of 0.2mm are machined using a CNC drilling machine. Chemical copper plating and electroplating are then performed. The copper thickness of the hole wall is ≥25μm. Blind buried holes are used for critical signal holes. Step 4: Component soldering and assembly. Surface components are soldered using SMT (Surface Mount Technology). The reflow soldering temperature profile includes the preheating zone, heating zone, peak zone, and cooling zone. After the core chip is soldered, bottom filling is performed, and automotive-grade epoxy resin is cured at 150°C for 60 minutes. Step 5: Strengthening and testing. Apply a 20-30μm thick conformal coating to the non-soldering areas of the circuit board. After curing, conduct electrical performance tests, environmental stress tests, and functional verification tests in sequence.

2. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The reflow soldering temperature profile in step four is as follows: Preheating zone: temperature 150-180℃, time 60s; Heating zone: temperature 180-220℃, time 30s; Peak zone: temperature 245±5℃, time 10s; Cooling zone: temperature less than 100℃, time 60s.

3. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The electrical performance test in step five uses flying probe testing with 100% coverage. The environmental stress test includes 50 cycles of high and low temperature cycling and 20 hours of vibration testing. The functional verification test includes connecting to the actual vehicle vision system to conduct at least 1,000 target recognition and control response tests.

4. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The method also includes wiring design and optimization before step two, specifically including: Based on the intelligent vehicle's sensing, decision-making, and control functions, a software and hardware interface protocol is defined and input to the integrated cabling module for verification. Based on the wiring information of the two-sided circuit board, the signal layer connection relationship is defined, the package stacking method is optimized, and a single-layer layout is generated; The layout editing module enables two-layer interactive design of the layout, generating the final wiring information.

5. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 4, characterized in that, In the cabling design, the interface protocol includes at least one of CAN FD, Ethernet AVB, and GMSL2, and a signal equalizer and EMC shielding circuit are used to ensure that the signal transmission bit error rate is ≤ .

6. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The method also includes short-circuit withstand level testing of the circuit board, specifically including: Metal conductive particles of different diameters are attached to the surface of the copper layer as short-circuit points; By simulating the particle sliding process and using SEM to characterize the surface state of the copper layer, the short-circuit withstand level of the circuit board can be determined.

7. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The method also includes conducting vibration resistance level tests on the circuit board, specifically including: Vibration experimental parameters were constructed, and the stress distribution at the interface between the copper layer and the dielectric layer at different vibration frequencies was characterized by SEM. The shear resistance rating of the circuit board was determined by combining conductivity tests.

8. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 7, characterized in that, The conductivity test specifically includes: The circuit board is divided into equilateral triangular regions with side lengths of 5 to 10 cm; Use a multimeter to measure the conductivity between the vertices and center of the triangle; The shear resistance level of the circuit board is determined based on the distribution of conductivity data.

9. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The method further includes impedance consistency testing of the circuit board, specifically including: Measure the conductivity of each signal layer; Calculate the average impedance, impedance non-uniformity, and impedance coefficient of variation; Inspect the diameter and accuracy of vias, the coplanarity of copper layers, and the flatness of dielectric layers.

10. The method for fabricating a vision-based intelligent vehicle circuit board according to claim 1, characterized in that, The circuit board includes a core control unit, a visual signal interface module, a multi-source fusion and control output unit, an environmental adaptive protection unit, and an expandable storage and upgrade unit. The core control unit adopts a dual-core architecture of main computing SoC and security MCU. The main SoC is an automotive-grade NVIDIA Jetson AGX Orin chip, and the security MCU is an Infineon AURIX TC49x series chip. The visual signal interface module integrates 4 GMSL2 interfaces and 2 Ethernet AVB interfaces, and has a built-in signal equalizer and EMC shielding circuit.

Images