System and method for eliminating blind area of A column of vehicle

Abstract

The embodiment of the invention discloses a system and method for eliminating a blind area of a vehicle A column, relates to the technical field of vehicle control, and aims to solve the problem that the blind area of the vehicle A column is difficult to adaptively and dynamically eliminate. The system comprises a multi-mode image fusion system which is integrated on a left A column and a right A column of a vehicle, comprises an optical reflection module and an electronic display module, and is used for superposing information of each module to generate composite view content; the dynamic intelligent control system is integrated with the Internet of Vehicles, comprises a plurality of sensing modules and is used for receiving the multi-source sensing signals and dynamically adjusting the composite view content based on the multi-source sensing signals; and the active safety system is used for acquiring the adjusted composite view content, identifying the A-column blind area risk characteristics of the vehicle, and performing graded safety response based on the A-column blind area risk characteristics.

Description

Technical Field

[0001] This specification relates to the field of vehicle control technology, and in particular to a system and method for eliminating the blind spot of the A-pillar of a vehicle. Background Technology

[0002] The A-pillar, located between the engine compartment and the passenger compartment, above the side mirrors, is the pillar on the left and right sides of a car's windshield. Its primary function is to connect the roof to the body and protect the passenger compartment in a collision. However, as a core load-bearing component, the A-pillar's width obstructs part of the driver's turning view, creating a blind spot. This blind spot expands exponentially with vehicle speed, making it difficult for the driver to perceive sudden road conditions and the external driving environment in real time. This significantly increases the risk of collisions to pedestrians or non-motorized vehicles. In dynamic driving scenarios, the interaction between the blind spot and the vehicle's trajectory further exacerbates driving hazards.

[0003] Current measures to eliminate the impact of the A-pillar blind spot mainly focus on driving behavior guidelines and local structural improvements. However, driving behavior guidelines such as walking around the vehicle for inspection, adjusting rearview mirrors, and actively observing with sensors when turning rely on the driver's observation and are therefore subject to delays. Local structural improvements, such as the design of the A-pillar triangular window, directly conflict with the goal of reducing the blind spot, and since blind spots vary across different vehicle models, there is a lack of universally applicable solutions for A-pillar blind spots.

[0004] Therefore, how to adaptively and dynamically eliminate the A-pillar blind spot without compromising the safety of the vehicle body structure has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of the above problems, this application provides a system and method for eliminating the blind spot of the A-pillar of a vehicle to overcome or at least partially solve the above problems. The technical solution is as follows: A system for eliminating blind spots caused by the A-pillar of a vehicle, applied to a vehicle, the system comprising: a multimodal image fusion system, a dynamic intelligent control system, and an active safety system; The multimodal image fusion system is integrated into the left and right A-pillars of the vehicle and includes an optical reflection module and an electronic display module, used to overlay information from each module to generate composite visual content. The dynamic intelligent control system is integrated with the Internet of Vehicles and includes multiple sensing modules for receiving multi-source sensing signals and dynamically adjusting the composite field of view content based on the multi-source sensing signals. The active safety system is used to acquire the adjusted composite field of view, identify the risk characteristics of the vehicle's A-pillar blind spot, and perform graded safety responses based on the risk characteristics of the A-pillar blind spot.

[0006] The multimodal image fusion system integrates optical reflection and electronic display modules. By superimposing the images from each module, it generates a composite field of view, achieving blind spot visualization without altering the A-pillar's load-bearing structure. The dynamic intelligent control system receives multi-source perception signals to dynamically adjust the content of the composite field of view. Based on the adjusted composite field of view, the active safety system performs risk identification and graded response, achieving dynamic elimination of the vehicle's A-pillar blind spot and active safety protection. This solves the problems of incomplete field of view compensation and lack of adaptability and universality of traditional single optical or electronic solutions, while also overcoming the inability of fixed field-of-view assistance to cope with dynamic driving scenarios and real-time risk intervention, thus improving vehicle safety in complex traffic environments.

[0007] Optionally, the optical reflection module includes a set of reflective mirrors respectively disposed on the inner side of the left and right A-pillars for reflecting the optical image in the blind zone; The electronic display module includes a transparent display screen integrated on the surface of the A-pillar, used to receive real-time blind spot images acquired by a miniature wide-angle camera on the outside of the A-pillar, and to overlay the images with the blind spot optical images.

[0008] By placing reflector groups on the inner sides of the left and right A-pillars of the vehicle to reflect the actual blind spot optical image, and simultaneously using a transparent display screen integrated into the A-pillar surface to receive and display the real-time blind spot image acquired by a miniature wide-angle camera, this composite field of view is generated by superimposing the optical image with the real blind spot optical image. This retains the optical reflection image, which conforms to human visual habits, while using the electronic display image to supplement the shortcomings of optical reflection in terms of viewing angle width, low-light detail, and information enhancement. The physical superposition of the two images through the transparent display screen allows the driver to simultaneously receive the real optical path information of the blind spot optical image and the enhanced information of the real-time blind spot image. This provides a continuous blind spot field of view without switching visual focus, effectively reducing cognitive load and improving information acquisition efficiency.

[0009] Optionally, the optical reflection module includes: a first optical reflection group and a second optical reflection group symmetrically arranged on the inner sides of the left and right A-pillars based on a sliding rail mounting structure; Each of the optical reflection groups includes: A reflector, wherein the reflector is an asymmetric curvature reflector, is used to acquire images of the blind spot of the opposite A-pillar; A presentation mirror, located below the reflector, is connected to the corresponding A-pillar via an adjustment device in the sliding rail mounting structure, and is used to present the light from the reflector to the driver's field of vision.

[0010] The sliding rail structure allows the optical reflection module to support optical path calibration based on the characteristics and habits of different drivers, ensuring that reflected images are presented in a personalized manner within the field of vision of each driver. Furthermore, the asymmetric curvature reflector specifically expands the effective field of view and reduces image distortion, improving the clarity and realism of the blind spot optical images acquired by the optical reflection module, thus providing high-quality blind spot optical images for subsequent processing.

[0011] Optionally, the electronic display module further includes: a microprism array for driving the transparent display screen; The microprism array is fabricated on the light-emitting layer of each sub-pixel of the transparent display screen. It is used to refract the display light reflected by the light-emitting layer to adjust the propagation direction of the display light. This allows the real-time blind spot image generated by the transparent display screen based on the display light to be spatially fused with the blind spot optical image to obtain composite visual content.

[0012] By fabricating microprism arrays on the luminescent layer of each sub-pixel of a transparent display screen, the reflected light from the luminescent layer can be refracted in a directional manner using these microprism arrays. This adjusts the direction of light propagation, enabling precise matching and fusion of real-time blind zone images and actual blind zone optical images in terms of spatial position and viewing angle. This solves the problem of parallax misalignment and difficulty in natural superposition between electronic and optical images caused by light scattering in traditional transparent displays. Furthermore, the spatial fusion of real-time blind zone images and actual blind zone optical images enhances the spatial coherence and visual realism of the composite field of view.

[0013] Optionally, the real-time blind spot image generated based on the displayed light is spatially fused with the blind spot optical image to obtain composite visual content, specifically including: Based on the spatial pose relationship between the optical reflection module and the transparent display screen, establish the pixel coordinates of the real-time blind spot image and the mapping relationship between them and the projection optical path where the blind spot optical image is located; The real-time blind spot image is subjected to perspective correction according to the mapping relationship, so as to align the real-time blind spot image with the blind spot optical image; The aligned real-time blind spot image and the blind spot optical image are pixel-wise superimposed to generate the composite field of view content on the transparent display screen.

[0014] A mapping relationship was established between the pixel coordinates of the real-time blind zone image and the projection optical path of the blind zone optical image. Based on this mapping relationship, perspective correction was performed on the real-time blind zone image, and it was aligned and superimposed with the blind zone optical image, realizing the fusion of the real-time blind zone image and the blind zone optical image to obtain composite field-of-view content. Perspective correction effectively eliminated image distortion caused by the difference between the camera's viewing angle and the optical reflection path, and the pixel superposition generated composite field-of-view content significantly improved the intuitiveness of the blind zone information presentation.

[0015] Optionally, receiving multi-source sensing signals and dynamically adjusting the composite field of view content based on the multi-source sensing signals specifically includes: Receive multi-source perception signals to determine the current driving scenario and driver intention based on the multi-source perception signals, and match the vision adjustment priority based on the current driving scenario and driver intention to determine the vision compensation area and compensation data corresponding to the A-pillar blind spot of the vehicle. Based on the field of view compensation area and compensation data, an adjustment instruction is generated; The electronic display module is adjusted based on the adjustment command, and / or the optical path direction of the optical reflection module is adjusted to obtain the adjusted composite field of view content.

[0016] By receiving multi-source sensing signals to determine the driving scenario and driver intentions, and matching the priority of vision adjustment according to the driving scenario and driver intentions, and determining compensation data, the system generates adjustment commands to control the electronic display and / or optical reflection module, realizing dynamic intelligent adjustment of composite vision content. This ensures that critical blind spot information is provided to the driver at critical moments, transforming the static A-pillar blind spot elimination method into a dynamic A-pillar blind spot elimination method, improving the effectiveness of information presentation, helping to reduce driver cognitive load, and enhancing driving safety.

[0017] Optionally, receiving multi-source perception signals to determine the current driving scenario and driver intention based on the multi-source perception signals, and matching the field of vision adjustment priority based on the current driving scenario and the driver intention to determine the field of vision compensation area and compensation data corresponding to the A-pillar blind spot of the vehicle, specifically including: Based on the multi-source perception data and preset thresholds, the current driving scenario and driver intent are determined; A preset priority mapping rule library is invoked to assign basic priority weights to the driving scenario model and the driver intention model. Based on the dynamic arbitration result of the current driving scenario and the driver intention, the basic priority weights are corrected to obtain the field of view adjustment priority. Based on the field of view adjustment priority, the field of view compensation area for the A-pillar blind spot on different priority sides is determined, and compensation data including electronic display enhancement parameters and optical reflection adjustment parameters is generated.

[0018] Multi-source perception data is compared with preset thresholds to identify driving scenarios and driver intentions. After obtaining basic weights, dynamic arbitration is used to adjust the weights and determine the priority of vision adjustment. This approach establishes a quantitative identification and weight mapping foundation for scenarios and intentions through preset thresholds and a rule base, ensuring the systematic and predictable nature of decision-making. Furthermore, the introduction of a dynamic arbitration mechanism allows for real-time comprehensive evaluation and weight adjustment of potentially conflicting or overlapping driving scenarios and driver intentions, effectively avoiding decision-making biases caused by fixed rules. Determining the vision compensation area and parameters based on the vision adjustment priority helps adjust for complex vision content in real-world driving environments, improving the scenario adaptability of A-pillar blind spot elimination.

[0019] Optionally, the adjusted composite field of view content is acquired, the risk characteristics of the vehicle's A-pillar blind spot are identified, and a graded safety response is performed based on the risk characteristics of the A-pillar blind spot, specifically including: Based on the adjusted composite field of view content, risk targets in the target detection area are identified; wherein, the target detection area is the projection area corresponding to the A-pillar blind spot; Feature extraction is performed on the risk target to obtain the risk characteristics of the vehicle's A-pillar blind spot; Based on the multi-source perception data, the current driving condition of the vehicle is determined. Based on the current driving condition and the risk characteristics, a risk assessment is performed on the risk target to obtain the risk level of the risk target. Trigger and execute graded security response operations corresponding to the risk level, continuously monitor the risk of the risk target, and adjust the composite view content and graded security response operations accordingly.

[0020] Based on the optimized composite field-of-view content recognition blind spot risk target, and through feature extraction of this risk template, a reliable basis for risk assessment is provided. By combining risk features with real-time driving conditions for dynamic risk determination, the assessment results more closely reflect the degree of danger in actual driving scenarios. Graded safety response based on risk levels enables timely intervention in sudden blind spot risks, enhancing driving safety and reducing accident risk.

[0021] Optionally, the sensing module specifically includes: An eye-tracking module includes a camera located inside the vehicle, used to acquire images of the vehicle driver and determine the driver's head position and gaze direction based on the acquired images. The vehicle status perception module, integrated with the vehicle network, is used to acquire the vehicle's dynamic data via the vehicle CAN bus. An environmental perception module is used to acquire external environmental data of the vehicle.

[0022] By using an eye-tracking module to capture the driver's head position and gaze direction in real time, the system can accurately understand the driver's attention distribution and observation intentions. The vehicle state perception module acquires dynamic vehicle data, helping to ensure the timeliness and accuracy of scene recognition and decision-making responses. The environmental perception module acquires data on the external environment. The multi-sensor module enables the acquisition of multi-source perception data of the vehicle, providing a reliable data foundation for A-pillar blind spot elimination.

[0023] A control method for eliminating vehicle A-pillar blind spot system, wherein the A-pillar blind spot elimination system is any one of the above-mentioned vehicle A-pillar blind spot elimination systems; The control method for eliminating the blind spot of the vehicle's A-pillar specifically includes: The multimodal image fusion system overlays information from each module to generate composite visual content; wherein, the multimodal image fusion system is integrated into the left and right A-pillars of the vehicle and includes an optical reflection module and an electronic display module; The dynamic intelligent control system receives multi-source sensing signals from multiple sensing modules and dynamically adjusts the composite field of view content based on the multi-source sensing signals. Based on the adjusted composite field of view obtained by the active safety system, the risk characteristics of the vehicle's A-pillar blind spot are identified, and a graded safety response is performed based on the risk characteristics of the A-pillar blind spot.

[0024] By employing the aforementioned technical solution, this application provides a system and method for eliminating blind spots in vehicle A-pillars. The multimodal image fusion system integrates optical reflection and electronic display modules, generating a composite field of view through the superposition of images from each module. This achieves blind spot visualization without altering the load-bearing structure of the A-pillar. A dynamic intelligent control system receives multi-source sensing signals to dynamically adjust the content of the composite field of view and utilizes an active safety system to perform risk identification and graded response based on the adjusted composite field of view. This achieves dynamic elimination of the vehicle's A-pillar blind spot and active safety protection. It solves the problems of incomplete field of view compensation and lack of adaptability and universality associated with traditional single optical or electronic solutions, while also overcoming the inability of fixed-view assistance to cope with dynamic driving scenarios and real-time risk intervention, thus improving vehicle safety in complex traffic environments. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1This is a schematic diagram of a system for eliminating the blind spot of a vehicle's A-pillar, provided as an embodiment of this specification. Figure 2 This is a schematic flowchart illustrating a method for eliminating the blind spot of a vehicle's A-pillar, as provided in an embodiment of this specification. Detailed Implementation

[0026] This specification provides a system and method for eliminating the blind spot of the A-pillar in a vehicle.

[0027] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0028] The A-pillar, located on the left and right sides of the windshield of a car, lies between the engine compartment and the passenger compartment, above the side mirrors. Its primary function is to connect the roof to the body and protect the passenger compartment in a collision. However, as a core load-bearing component, the width of the A-pillar can obstruct the driver's turning view, creating a blind spot. In practical applications, the width of the A-pillar in ordinary sedans is generally 8-12 centimeters, while some models have A-pillars exceeding 15 centimeters, creating a 6-8 degree fan-shaped blind spot. When the vehicle turns at a 45-degree angle, the A-pillar can completely obstruct pedestrians under 1.7 meters tall or non-motorized vehicles within 0.8 meters in width. Especially at speeds of 40-60 km / h, the blind spot expands exponentially with speed, making it difficult for the driver to perceive sudden road conditions and the external driving environment, significantly increasing the risk of collisions to pedestrians or non-motorized vehicles.

[0029] Furthermore, in dynamic driving scenarios, the interaction between blind spots and vehicle trajectory exacerbates driving hazards. Specifically, when turning left, the blind spot area of ​​the left A-pillar is significantly larger than that of the right, and multiple dimensions of information such as oncoming traffic and pedestrians must be considered. Traditional methods that rely on the driver to adjust their perspective by moving their body have limited response efficiency. Especially in low-light environments and sudden road conditions, human observation is delayed and cannot cover the dynamic characteristics of blind spot changes.

[0030] Currently, the main methods for addressing the A-pillar blind spot are through drivers observing the external driving environment according to driving behavior guidelines and through improvements to the vehicle's local structure. However, human observation has a time lag and cannot cover the dynamic characteristics of blind spot changes. Furthermore, there is a lack of universal solutions for the differences in blind spots across different vehicle models. Moreover, there is a direct conflict between the current demands for strengthening the A-pillar structure and reducing the blind spot. Engineering design requires trade-offs between vehicle body rigidity, collision safety, and visibility, making it difficult to eliminate the blind spot through simple structural adjustments to the existing A-pillar. Intelligent driver assistance technologies such as blind spot monitoring cameras and millimeter-wave radar are not yet widely adopted in traditional vehicles, and relying on expensive high-end solutions cannot meet the needs of the entire market.

[0031] Therefore, how to adaptively and dynamically eliminate the A-pillar blind spot without compromising the safety of the vehicle body structure has become a technical problem that urgently needs to be solved by those skilled in the art.

[0032] One embodiment of this application provides a system for eliminating the blind spot of a vehicle's A-pillar, such as... Figure 1 As shown, Figure 1 This application provides a system for eliminating the A-pillar blind spot in vehicles. The system, applied to vehicles, includes a multimodal image fusion system 100, a dynamic intelligent control system 101, and an active safety system 102. Through the collaborative operation of the multimodal image fusion system 100, the dynamic intelligent control system 101, and the active safety system 102, optical transmission, electronic display, environmental perception, intelligent decision-making, and vehicle control are integrated across domains. This not only dynamically eliminates the A-pillar blind spot physically, resolving the conflict between structural safety and a wide field of vision, but also, through intelligent information enhancement and graded safety response, achieves a transition from passively adapting to blind spots to actively warning of blind spots, thereby improving vehicle safety on roads.

[0033] Specifically, as mentioned in the background, traditional A-pillar blind spot solutions for vehicles, while improving the field of vision to a limited extent by reducing the A-pillar cross-section or adding triangular windows, compromise the vehicle's passive safety by weakening the A-pillar's core load-bearing and energy-absorbing capabilities in rollovers and collisions. In other words, there is a direct conflict between overall safety and eliminating blind spots. Auxiliary mirrors / displays that rely on driver behavior or fixed angles suffer from problems such as reaction lag, fixed viewing angles, and distraction. This results in constantly changing blind spot ranges and risk focal points during dynamic scenarios like turning and lane changes, which static solutions cannot provide real-time adaptive vision compensation. Furthermore, simple electronic display solutions at the A-pillar provide a video view independent of the real road environment. Drivers need to switch and spatially transform between the real world and the video view to establish a correspondence. This information separation increases the complexity and time cost of judgment in emergency situations, potentially delaying evasive reactions. Based on this, in this embodiment, the multimodal image fusion system is integrated into the left and right A-pillars of the vehicle, including an optical reflection module and an electronic display module, for superimposing information from each module, that is, spatially fusing information from two different modalities, physical optical imaging and digital electronic display, to generate composite visual content.

[0034] The optical reflection module includes mirror groups respectively set on the inner side of the left and right A-pillars to reflect the optical image of the blind spot. Its working principle is to use the law of light reflection to directly reflect the real light of the blind spot on the opposite side that is blocked by the A-pillar on the same side of the driver through a preset high-precision optical path, so as to use the optical image of the blind spot as the basis image for subsequent visual fusion.

[0035] The electronic display module is also integrated inside the A-pillar, specifically as a flexible transparent display screen covering the inner surface of the A-pillar. This module receives real-time blind spot video streams and real-time blind spot images captured by a miniature wide-angle camera on the outer side of the A-pillar, and overlays these images with the blind spot optical image to generate composite visual content. This electronic display module not only displays the real-time blind spot image but also uses image fusion algorithms to perform real-time spatial registration and pixel-level fusion of the real-time blind spot image generated by the electronic display module and the blind spot optical image generated by the optical reflection module at the driver's viewpoint. For example, it uses deep learning algorithms to generate distortion-free composite images, ensuring image continuity in dynamic scenes and obtaining composite visual content. This process transforms the two independent images—the blind spot optical image and the real-time blind spot image—into a single, coherent composite visual content.

[0036] The transparent display screen is an organic light-emitting diode (OLED) display screen, which serves as a shared fusion plane for optical and electronic images and the final visual output interface.

[0037] The aforementioned multimodal image fusion system 100, without altering the original structure of the A-pillar, superimposes the blind spot optical image acquired by the optical reflection module and the real-time blind spot image acquired by the electronic display module. This engineering approach separates the load-bearing structure from the field of vision function, eliminating the A-pillar blind spot without sacrificing collision safety. The composite field of vision content generated by the electronic display module on the transparent screen ensures that the image displayed on the left A-pillar screen visually overlaps with the actual physical image of the bicycle on the right side, projected by the optical system and seen by the driver through the same screen. This eliminates the need for the driver to first look at the screen and then turn their head or guess the actual position, shortening the driver's reaction time in complex road conditions.

[0038] When displaying reflected images on a transparent screen of an electronic display module based on the aforementioned multimodal image fusion system, if only a fixed, universal blind spot video stream or reflected image is provided, this fixed-angle blind spot reflected image, in some dynamic driving scenarios, such as high-speed straight-line cruising where the driver does not need to pay attention to the blind spot, will become an irrelevant visual interference, distracting the driver from the main road conditions ahead. Furthermore, in situations requiring attention to the blind spot, such as turning or changing lanes, the fixed-angle display method emphasizes high-risk targets or areas within the blind spot, making it difficult for the driver to quickly capture risk information in the regular view. Moreover, drivers of different heights, or even the same driver in different seating positions, have different viewpoints relative to the A-pillar display. A fixed-angle optical reflection path and display image cannot provide all users with an optimal, distortion-free field of view, potentially leading to image distortion or information misalignment, affecting user experience and reliability. In environments with drastic changes in lighting, such as at night, in tunnels, or in rain and fog, a fixed display brightness can also cause nighttime glare or poor visibility during the day. Therefore, in order to enable the composite field of view content generated by the aforementioned multimodal image fusion system 100 to adapt to dynamic driving scenarios, achieve personalized adaptation between the display and the driver, and improve the system's universality across different vehicle models, the system for eliminating the vehicle's A-pillar blind spot also includes: a dynamic intelligent control system 101.

[0039] The dynamic intelligent control system 101 is integrated with the vehicle network and includes multiple sensing modules to receive multi-source sensing signals. These sensing modules specifically include: an eye-tracking module, a vehicle status sensing module, and an environmental sensing module. The eye-tracking module includes a camera located inside the vehicle to acquire images of the driver and determine the driver's head position and gaze direction based on these images. Specifically, image processing algorithms analyze images acquired by in-vehicle cameras, such as infrared cameras, to identify the driver's eye features and movement trajectories, thereby determining the driver's gaze direction. The vehicle status sensing module, integrated with the vehicle network, acquires vehicle dynamic data via the vehicle's CAN bus. Specifically, it acquires vehicle dynamic data, including steering angle, vehicle speed, and navigation information, via the CAN bus to predict the driver's field of vision needs. The environmental sensing module acquires data about the vehicle's external environment. Specifically, the environmental perception module integrates a light sensor and a rain sensor to monitor the ambient light intensity and rainfall in real time, so as to automatically adjust the display brightness and contrast. For example, it can activate a low blue light mode at night to reduce screen brightness and reduce eye strain on the driver; and increase screen brightness and contrast in bright daylight to ensure that the image is clearly visible.

[0040] The dynamic intelligent control system 101 analyzes received signals such as vehicle steering intention, driver's line of sight, and changes in ambient light in real time. Based on the analysis of multi-source sensing signals, it generates control commands and dynamically adjusts the composite field of vision content. For example, when the vehicle turns left, the system can automatically enhance the contrast of the blind spot image in the left A-pillar display; when it detects that the driver is looking at the right A-pillar area, it can fine-tune the angle of the left optical reflector to optimize the center of the field of vision, thereby improving the scene adaptability of the vehicle's A-pillar blind spot elimination.

[0041] This dynamic intelligent control system 101, by introducing dynamic intelligent control based on multi-source sensing signals, can intelligently judge driving scenarios and adjust the content of the complex field of vision. For example, when the vehicle is traveling straight and at a stable speed, it automatically reduces the brightness of the blind spot image or switches it to simplified speed and navigation information to reduce interference with the driver's primary task. If the steering wheel angle or turn signal indicates an imminent left turn, the system instantly increases the contrast of the blind spot image on the left A-pillar display to the highest level before the turn occurs and highlights moving objects, effectively avoiding information overload and fatigue for the driver. Through eye tracking, the system obtains the driver's head position and gaze direction in real time. Based on this data, the system dynamically calculates and controls the micro servo motor to fine-tune the angle of the optical reflector, ensuring that the center line of the reflected light path is precisely aligned with the driver's field of vision. This allows drivers of different heights and sitting postures to obtain an image with a direct view and no geometric distortion. Furthermore, the system can adjust the brightness and color temperature of the transparent display screen through the environmental perception module to ensure that the content of the complex field of vision is clearly distinguishable under any lighting conditions.

[0042] The vehicle-to-everything (V2X) network can be integrated with V2X modules such as 5G to obtain the location information of other vehicles and pedestrians at the intersection where the vehicle is located. This information is then used to perform spatiotemporal calibration with the composite visual content of a multimodal image fusion system, highlighting potential risk targets on the screen. It can also be linked with ADAS systems. When blind-spot radar detects pedestrians or obstacles, a red warning frame is overlaid on the electronic screen with haptic feedback. Alternatively, virtual lane lines, traffic sign recognition results, and real-time traffic information can be projected onto a transparent screen, supporting gesture control to switch display modes.

[0043] After achieving dynamic visual elimination and intelligent information presentation of the A-pillar blind spot, drivers can see the risk targets in the blind spot. However, in complex dynamic traffic environments, drivers may misjudge risk targets due to multiple factors, resulting in a decision-making speed lagging behind the risk development speed. Furthermore, current blind spot monitoring systems generally only provide audible and visual warnings. When the driver fails to respond to this warning, or when the risk escalates rapidly in a very short time, it is difficult to coordinate vehicle braking, steering, and other actuators to achieve timely driving safety protection. Therefore, in order to achieve timely safety response after obtaining and displaying composite visual content based on the multimodal image fusion system 100 and dynamic intelligent control system 101, the active safety system 102 continuously analyzes the dynamically adjusted composite visual content, uses computer vision algorithms to identify potential risk targets in the blind spot such as pedestrians and non-motorized vehicles, and extracts their risk characteristics such as relative distance, speed, and trajectory. Based on the severity of the risk characteristics, the system activates a graded safety response mechanism. This tiered safety response mechanism can range from overlaying visual warnings and issuing audible alerts on a composite field of vision, to providing auxiliary braking or steering torque when the driver's reaction is insufficient, and even actively taking over the vehicle to avoid obstacles in emergency situations.

[0044] In a specific application scenario, the first-level response to the detection of potential collision risks or lane departure trends initiates a non-intrusive warning. A 3D warning image is projected onto the windshield via a Head-Up Display (HUD), accompanied by directional tactile alerts from the seat vibration module. Simultaneously, a high-frequency speaker in the steering wheel emits a simulated ambient sound warning. This first-level response eliminates blind spots caused by the A-pillar, allowing the system to monitor without intervention and preserving driver control. For the second-level response, where the driver fails to react effectively within a set threshold (e.g., 2.5 seconds), the drive and braking domains are linked for control. The seatbelt is pre-tensioned to eliminate redundant gaps, braking force is applied in stages, and a lateral avoidance path is calculated simultaneously. After acquiring the trajectories of surrounding vehicles via the vehicle-to-everything (V2X) network, the steering system enters a quasi-responsive mode. Only when the driver's steering wheel input is insufficient does the electric power steering (EPS) system provide complementary torque to correct the path. For collisions lasting less than 0.8 seconds or when the Driver Monitoring System (DMS) determines the driver is incapacitated (Level 3 response), the system enters fully autonomous control mode. In this mode, the powertrain performs fuel and power cut-off, the IBooster hydraulic unit establishes maximum braking pressure, and lateral control employs a path replanning strategy. This involves using independent braking of all four wheels to create lateral vector torque, thus deflecting the vehicle off its intended path. Furthermore, it should be noted that the main control chip in this A-pillar blind spot elimination system is an automotive-grade System-on-Chip (SoC), supporting multiple video inputs and real-time image processing, with a power consumption of less than 5W.

[0045] By continuously analyzing the composite field of view content during this process, not only can the risk targets in the A-pillar blind spot be identified, but also the risk characteristics of the risk targets, such as collision time, expected collision point, and motion vector, can be determined. Based on the risk characteristics, the corresponding graded safety response can be determined, thereby improving the accident avoidance rate in driving scenarios.

[0046] Furthermore, in the aforementioned multimodal image fusion system 100100, the optical reflection path of the optical reflection module is a fixed physical optical path, while the transparent display screen of the electronic display module is an independent planar display device. Both have manufacturing and assembly tolerances in their installation positions and angles within the vehicle's A-pillar. Therefore, without calibration, the electronic image and the optical image will appear in different spatial positions and perspective depths to the driver, leading to severe ghosting and misalignment problems. This not only fails to eliminate blind spots but also causes visual confusion and misjudgment. Moreover, the blind spot optical image generated by the reflector of the optical reflection module has a specific viewing angle and optical distortion. The real-time blind spot image acquired by the wide-angle camera of the electronic display module also has its own lens distortion and perspective relationship. If the unprocessed electronic image of the real-time blind spot is directly displayed on the screen, it will be difficult to achieve visual consistency and uniformity with the blind spot optical image.

[0047] Therefore, to solve this problem, in one embodiment, the optical reflection module includes a first optical reflection group and a second optical reflection group symmetrically arranged on the inner sides of the left and right A-pillars based on a sliding rail mounting structure. Each optical reflection group includes a reflector and a presentation mirror. The first optical reflection group has a first reflector, and the second optical reflection group has a second reflector. The first reflector is responsible for capturing the image in the blind area of ​​the right A-pillar and transmitting it to the first presentation mirror of the first optical reflection group through reflection. The second reflector is responsible for capturing the image in the blind area of ​​the left A-pillar and transmitting it to the second presentation mirror of the second optical reflection group through reflection. Furthermore, traditional plane mirrors or symmetrical curved mirrors will produce image distortion and edge blurring when reflecting and compressing images from long distances and large angles into the limited space of the A-pillar blind area. Therefore, the asymmetric curvature reflector used in this embodiment can match the curved surface of the A-pillar structure in a certain application scenario. After the reflected light is split by the prism, it covers the blind zone range of 6.8° on the left and 2.3° on the right, and the error is controlled within ±0.2°, thereby effectively eliminating the problem of overlapping views of traditional dual mirrors and directly projecting the light from the opposite blind zone into the driver's field of vision.

[0048] The projection mirror, located below the reflector, is connected to the corresponding A-pillar via an adjustment mechanism in a sliding mounting structure. It directs the light from the reflector into the driver's field of vision. Essentially, light from the blind spot near the opposite A-pillar is first captured and reflected by the upper reflector. The light then travels downwards, is received by the projection mirror, and undergoes a final second reflection. This reflection then projects the optical image of the blind spot from the optical reflection module onto the transparent display screen of the electronic display module. At this point, the transparent display screen shows a real-time blind spot image from an external camera, thus creating a composite view by overlaying the blind spot optical image and the real-time blind spot image on the transparent screen.

[0049] It should be noted that this sliding rail mounting structure integrates a servo motor adjustment device. Upon first use, the motor drives the sliding rail to move the entire reflector assembly based on the driver's seat position and height, allowing for a one-time coarse adjustment so that the projection range of the image captured by the optical reflection module covers the driver's field of vision. During driving, combined with the dynamic intelligent control system 101, the device can adjust the reflectors or display mirrors in the first and second optical reflection groups based on real-time feedback from multi-source sensing signals. In a certain application scenario, the reflector and display module use a sliding rail mounting structure, supporting 0-15cm longitudinal adjustment to adapt to the A-pillar width of different vehicle models.

[0050] In this embodiment, the electronic display module further includes a microprism array that drives the transparent display screen. The microprism array is located above the light-emitting layer of each sub-pixel of the transparent display screen and is fabricated using photolithography to create a micro-transparent display screen array. Its function is to change the direction of light propagation, causing the light emitted from the light-emitting layer to be refracted directly at the interface of the transparent display screen, thus facilitating better control and focusing of the light. By adjusting the propagation direction of the display light reflected from the light-emitting layer through refraction, the real-time blind zone image generated by the transparent display screen based on the display light is spatially fused with the blind zone optical image to obtain composite visual content.

[0051] This process solves the spatial misalignment problem between optical reflection images and electronic display images by introducing a microprism array, achieving spatial fusion of the two. Traditional superimposed displays are prone to image shift and ghosting, causing visual confusion for the driver. However, the microprism array, fabricated on the sub-pixel emitting layer of the transparent display, can precisely adjust the propagation direction of the electronic display light, ensuring a high degree of spatial matching between the real-time blind spot image generated by the electronic screen and the blind spot image of optical reflection. Simultaneously, the microprism array, fabricated on the sub-pixel emitting layer of the transparent display, can regulate the propagation direction of the electronic display light through its refraction control, and also improve the brightness uniformity and viewing angle range of the electronic image, ensuring that the driver can clearly see the fused composite visual content regardless of their seating position. The curved reflectors and microprism array are symmetrically arranged on the inner sides of the left and right A-pillars, and through high-precision reflection path design, the problem of overlapping viewing angles in traditional dual-mirror displays is eliminated.

[0052] Traditional image overlay methods often rely on simple image superposition, which is prone to image distortion due to differences in viewing angle. In this embodiment, however, based on the spatial pose relationship between the reflector, projection mirror, and transparent display screen in the aforementioned optical reflection module, a mapping relationship is established between the pixel coordinates of the real-time blind zone image and the projection optical path of the blind zone optical image. In one feasible embodiment, this spatial mapping relationship can be established by placing a calibration plate of known size at the blind zone location, capturing an image of the calibration plate with a camera, and simultaneously recording the projection of the optical system onto the calibration plate. A computer vision algorithm is then used to calculate the complete chain transformation relationship between the camera pixel coordinate system, the real-world coordinate system of the blind zone, and the virtual image point coordinates projected onto the display screen plane by the optical system—that is, the mapping relationship.

[0053] During driving, for each frame of real-time blind spot image from the electronic display module, the electronic display module of the multimodal image fusion system 100 uses the aforementioned mapping relationship to perform real-time image remapping on the real-time blind spot image. This converts the image, which is acquired from the perspective of the wide-angle camera and has its own lens distortion, into an image that should be seen from the driver's viewpoint through the current optical reflection system, thereby achieving pixel alignment between the real-time blind spot image and the blind spot optical image. The corrected and aligned real-time blind spot image is sent to the display buffer and displayed on the transparent display screen, while the blind spot optical image from the opposite blind spot passes through the transparent display screen. Since the real-time blind spot image has undergone perspective correction, the virtual information represented by each pixel corresponds spatially to the physical scene information carried by the real light rays of the blind spot optical image penetrating that pixel position. Therefore, the aligned real-time blind spot image and the blind spot optical image can be pixel-by-pixel superimposed to generate composite visual content on the transparent display screen.

[0054] The aforementioned sliding rail adjustment addresses individual differences, while the microprism array optimizes the display optical path. Algorithm correction achieves pixel-level alignment, and the combination of these three elements ensures a stable fused view for different drivers under various driving conditions. The use of asymmetric curvature mirrors and perspective correction ensures that the fused image conforms to the human eye's perspective expectations of the real world, avoiding distortion and discomfort and reducing cognitive load. The mapping relationship is based on physical calibration, independent of complex real-time calculations, providing a stable fusion benchmark for the system and ensuring long-term functional reliability.

[0055] After the multimodal image fusion system 100 generates composite visual content using its optical reflection module and electronic display module, it only displays it in a fixed and universal manner. While this static display scheme can achieve basic blind spot information presentation, its fixed display strategy makes it difficult to adapt to dynamic driving scenarios. Specifically, the risk of the vehicle's A-pillar blind spot is spatiotemporally specific. During high-speed straight-line cruising, continuously highlighting the blind spot image is an ineffective visual interference. However, in critical moments such as turning and lane changes, if the potential collision target within the blind spot cannot be immediately highlighted, the system for eliminating the vehicle's A-pillar blind spot is ineffective. Therefore, the display of this composite visual content is mismatched with dynamic risk scenarios. Furthermore, different drivers have vastly different optimal viewing points relative to the A-pillar display area due to differences in height, posture, and driving habits. A fixed optical reflection angle and electronic display perspective can cause geometric distortion or field-of-view deviation in the images seen by some users. In severe cases, this can cause the superimposed virtual information to be misaligned with the actual object, leading to misleading information and reducing credibility and universality. Furthermore, drastic changes in ambient lighting, such as when entering or exiting tunnels, encountering oncoming high beams at night, or experiencing rain or fog, can severely impact the visibility of transparent displays. Screens with fixed brightness will become difficult to see in strong light due to glare, while conversely, they will become too bright in low light, causing visual pollution. This transforms a system designed to eliminate blind spots near the A-pillar from a driving safety aid into a safety hazard.

[0056] As discussed above, if the field of view adjustment strategy is discrete and abrupt—meaning it only divides scenarios based on a single parameter such as vehicle speed or steering angle and matches a fixed adjustment scheme—then while it may provide basic blind spot visibility in some typical scenarios, the switching of the field of view might be achieved through sudden expansion or contraction of the display area, abrupt increase or decrease in screen brightness, or step-by-step adjustment of the reflector angle. Understandably, this discontinuous adjustment method disrupts the driver's visual continuity, causing visual discomfort and potentially leading to missed critical A-pillar blind spot risk information. Furthermore, if the field of view adjustment priority lacks a dynamic arbitration mechanism and is only adjusted based on preset weights, then when multiple driving scenario characteristics overlap, such as rain, curves, and driver line of sight deviation, a fixed priority will result in high-risk blind spots not being enhanced and non-critical areas being over-displayed. This advanced adaptation defect will be transmitted to the driver through visual interaction, thus affecting driving safety and operational convenience.

[0057] Based on this, in order to dynamically optimize the composite field of view content, in one embodiment, to ensure that not only can the field of view adjustment adapt to real-time complex scenes, but also to guarantee accurate priority matching and continuous smooth compensation parameters, this embodiment of the application receives multi-source perception signals, determines the current driving scene and driver intention based on the multi-source perception signals, and matches the field of view adjustment priority based on the current driving scene and the driver intention to determine the field of view compensation area and compensation data corresponding to the vehicle's A-pillar blind spot. The process is specifically as follows: The system receives multi-source sensing signals and compares the aforementioned multi-source sensing data with preset thresholds to construct a driving scenario model and a driver intent model. The preset thresholds are calibrated using extensive real-vehicle test data. For example, a turning angle >15°, vehicle speed <40km / h, and road curvature <200m are considered a curve scenario; the driver's gaze lingers on the left A-pillar blind spot for >1 second without activating the left turn signal, indicating the driver intends to focus on the left blind spot. Through cross-validation of multi-dimensional data, accurate judgment of the current driving scenario and driver intent is achieved, avoiding the bias of judgments based on a single parameter.

[0058] After obtaining the current driving scenario and driver intent, a preset priority mapping rule base is invoked. This rule base pre-stores the basic priority weights corresponding to different scenarios and intents. For example, the basic weight of the inner blind spot is 0.8 in a curve scenario, and the basic weights of the left and right blind spots are both 0.4 in a straight driving scenario; when the driver intends to focus on a certain side blind spot, the corresponding blind spot's basic weight increases by 0.3.

[0059] Based on the superimposed characteristics of the current driving scenario and the driver's intention, a dynamic arbitration strategy is used to correct the basic priority weights. This dynamic arbitration strategy aims to avoid the frequent failures of assigning priorities to driving scenarios and driver intentions solely based on preset fixed rules in real-world mixed dynamic scenarios, which could even introduce new decision-making risks. For example, when the vehicle activates the left turn signal due to driver error, indicating an active left turn, but the driver's gaze remains anxiously fixed on the right-side rearview mirror, indicating an active focus on the right, the fixed rules might incorrectly reinforce the left-side view while ignoring the potentially more pressing right-side risks perceived by the driver, leading to a mismatch between system output and actual safety requirements. Therefore, it is crucial to obtain the focus sides of the current driving scenario and the driver's intention; if these focus sides differ, a conflict exists. In this case, the driver's intention is quantified based on the multi-source perception data to obtain the driver's intention intensity. For example: Intention intensity = Driver's gaze dwell time × Gaze focus concentration. The urgency of the current driving scenario is also quantified based on the multi-source perception data, for example: Scenario urgency = Absolute value of steering wheel angle × Steering angular velocity. Furthermore, the driver's state is assessed using multi-source perception data to obtain the driver state confidence level. The obtained driver intent strength, scenario urgency, and driver state confidence are used as arbitration factors. Based on these factors, a pre-set weight correction function is executed to obtain the arbitration result. This pre-set weight correction function is set based on the goal of prioritizing the primary task defined by a clear scenario and increasing the priority of secondary tasks indicated by strong intent or objective risk. The specific weight correction function is not limited here. Based on the superimposed characteristics of the current driving scenario and driver intent, the basic priority weights are adjusted through dynamic arbitration—that is, weighted—solving the problem that traditional fixed weights cannot adapt to complex superimposed scenarios, ensuring that the priority of vision adjustment accurately matches the current core driving needs.

[0060] After obtaining the basic priority weights and the field of view adjustment priority, differentiated compensation can be performed on the A-pillar blind spots of different priorities based on the final output field of view adjustment priority. The field of view compensation area of ​​the A-pillar blind spot on different priority sides is determined, and compensation data including electronic display enhancement parameters and optical reflection adjustment parameters is generated.

[0061] Furthermore, in a feasible embodiment, before determining the current driving scenario and driver's intention based on the multi-source sensing signals, the dynamic intelligent control system 101 can preprocess the multi-source sensing signals after receiving them to obtain standardized multi-source sensing signals. That is, to ensure that multi-source sensing signals from different physical sources, with different dimensions, frequencies, and signal-to-noise ratios, can be processed uniformly and reliably, preprocessing and standardizing the multi-source sensing signals is necessary. Specifically, the timestamps of each multi-source sensing signal are obtained. Because the acquisition periods of data frames from different sensors are different—for example, CAN bus signals are 10ms or 20ms, camera images are 33ms / frame, and eye-tracking data can reach 60Hz—linear interpolation, spline interpolation, or buffer alignment strategies can be used to uniformly align all multi-source sensing signals to a preset processing time reference, such as the start time of the main control cycle of the dynamic intelligent control system 101. By synchronizing and aligning the multi-source sensing signals in time, the problem of inconsistent decision-making time caused by asynchronous signal transmission and processing delays is eliminated. After time alignment, the aligned multi-source sensing signals undergo data cleaning and outlier processing. This involves identifying and discarding values ​​that significantly exceed physical limits, and using sliding window mean filtering or median filtering to smooth short-term noise for signals susceptible to transient interference. This data cleaning and outlier processing of the aligned multi-source sensing signals improves data reliability and robustness. The cleaned multi-source sensing signals are then normalized and standardized to obtain standardized multi-source sensing signals, providing a reliable data foundation for the decision chain of the dynamic intelligent control system 101.

[0062] In a certain application scenario, receiving multi-source sensing signals and dynamically adjusting the composite field of view content based on the multi-source sensing signals can be achieved in the following ways: Based on factors such as vehicle driving status, ambient lighting conditions, and the driver's viewing angle, the system automatically adjusts the parameters of the optical and electronic systems to achieve optimal blind spot elimination. For example, when the vehicle is driving at night, the brightness of the display screen is automatically reduced to avoid glare for the driver; when the driver's viewing angle changes, the angles of the reflectors and the projection mirror are automatically adjusted to ensure that the blind spot image is always within the driver's field of vision.

[0063] Based on rain sensor signals, the camera module's self-cleaning mechanism is controlled. When rain is detected, the camera module's miniature air pump starts, periodically blowing air to remove dust from the lens; the electrostatic adsorption layer on the display panel surface reduces fingerprint residue, while the retractable cleaning brush bristles are driven by a miniature stepper motor to automatically perform cleaning actions, ensuring image clarity in inclement weather.

[0064] Based on eye-tracking results, the optical refraction angle of the display area and the projection range of the electronic screen are dynamically adjusted. For example, when the driver's gaze shifts to the left, the display ratio of the left blind spot image on the electronic screen is appropriately increased, allowing the driver to observe the situation in the left blind spot more clearly and reducing visual fatigue.

[0065] A miniature servo motor is integrated into the reflector and camera assembly. Based on feedback signals from driver eye tracking or head position sensors (such as infrared or ToF sensors), a control algorithm automatically adjusts the reflector angle and camera pitch angle to adapt to the field of view requirements of drivers of different heights. For example, for taller drivers, the reflector and camera angles are adjusted upwards to expand the field of view in the upper blind spot; for shorter drivers, the angles are adjusted downwards.

[0066] After achieving dynamic visual elimination of the A-pillar blind spot and presentation of composite visual content based on the above steps, early warning intervention is needed to ensure timely response to blind spot risks. Therefore, in one embodiment, to achieve timely response to risks in the A-pillar blind spot, a target risk area is identified within the projection area corresponding to the A-pillar blind spot. Based on the composite visual content, risk targets are identified only within this target risk area. For example, moving targets such as pedestrians and non-motorized vehicles in the target risk area are obtained through a target detection algorithm, while static, risk-free targets such as curbs and green belts are filtered out.

[0067] For identified risk targets, multi-dimensional feature extraction is performed to form a risk feature set. These features include target type, relative distance, relative speed, trajectory, and predicted collision time. Taking an electric bicycle appearing in a blind spot as an example, the system will extract risk features such as its non-motorized vehicle type, speed of 25 km / h, distance from the vehicle of 8 m, and direction of movement intersecting with the vehicle of 25 km / h's left-turn trajectory, to construct a complete risk feature profile.

[0068] Traditional systems typically rely on fixed physical thresholds for judgment, such as triggering an alarm if the distance is less than 5 meters. However, driving risks are inherently dynamic and relative. The level of danger represented by a pedestrian at the same distance differs depending on whether the vehicle is stationary, slowly creeping, traveling at high speed, or turning. Furthermore, the driver's response to the risk also affects the risk level. Static models cannot capture this dynamic relationship, leading to invalid alarms in safe scenarios. Therefore, after obtaining the risk characteristics of the A-pillar blind spot, the system simultaneously utilizes multi-source perception data provided by the dynamic intelligent control system 101 to determine the current driving condition. For example, the system determines whether the vehicle is in conditions such as "high-speed cruising," "traffic jam following," "turning left at an intersection," or "wet and slippery road conditions at night." Based on different conditions, the weights of each risk characteristic are dynamically adjusted to achieve risk assessment of the risk target and obtain its risk level. Based on the assessed risk level, the system triggers the tiered response operation mentioned in the above embodiment. This scenario-based risk assessment ensures the appropriateness of the response intensity, reduces false alarms and missed alarms, and improves driving safety.

[0069] Figure 2 This is a schematic flowchart of a method for eliminating the blind spot of the A-pillar of a vehicle, provided in an embodiment of this application.

[0070] For example, such as Figure 2 As shown, the A-pillar blind spot elimination system of this method is any of the above-mentioned vehicle A-pillar blind spot elimination systems; The control method for eliminating the blind spot of the vehicle's A-pillar specifically includes: S201: Generate composite visual content by superimposing information from each module through the multimodal image fusion system; wherein, the multimodal image fusion system is integrated into the left and right A-pillars of the vehicle and includes an optical reflection module and an electronic display module; S202: The dynamic intelligent control system receives multi-source sensing signals from multiple sensing modules and dynamically adjusts the composite field of view content based on the multi-source sensing signals. S203: Based on the adjusted composite field of view obtained by the active safety system, identify the risk characteristics of the vehicle's A-pillar blind spot, and perform a graded safety response based on the risk characteristics of the A-pillar blind spot.

[0071] It should be noted that all relevant content of each step involved in the above system embodiments can be referenced in the description of the corresponding method, and will not be repeated here.

[0072] The vehicle provided in this embodiment is used to execute the above-described braking force control method, and therefore can achieve the same effect as the above-described implementation method.

[0073] When using integrated units, the vehicle may include a processing module and a storage module. The processing module is used to control and manage the vehicle's actions. The storage module supports the vehicle in executing program code and data.

[0074] The processing module may be a processor or a controller, which can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0075] This embodiment also provides a computer-readable storage medium (including but not limited to disk storage, CD-ROM, optical storage, etc.) storing computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to realize a vehicle remote control method provided in the above embodiment.

[0076] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to realize a vehicle remote control method provided in the above embodiment.

[0077] The beneficial effects of the above embodiments can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0078] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0079] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0080] In the description of this disclosure, it should be understood that if the terms "upper", "lower", "front", "rear", "left" and "right" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the position or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this disclosure.

[0081] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0082] The above are merely embodiments of this disclosure and are not intended to limit the scope of this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims of this disclosure.

Claims

1. A system for eliminating the blind spot of the A-pillar in a vehicle, applied to a vehicle, characterized in that, The system includes: a multimodal image fusion system, a dynamic intelligent control system, and an active safety system; The multimodal image fusion system is integrated into the left and right A-pillars of the vehicle and includes an optical reflection module and an electronic display module, used to overlay information from each module to generate composite visual content. The dynamic intelligent control system is integrated with the Internet of Vehicles and includes multiple sensing modules for receiving multi-source sensing signals and dynamically adjusting the composite field of view content based on the multi-source sensing signals. The active safety system is used to acquire the adjusted composite field of view, identify the risk characteristics of the vehicle's A-pillar blind spot, and perform graded safety responses based on the risk characteristics of the A-pillar blind spot.

2. The system for eliminating the blind spot of the A-pillar of a vehicle according to claim 1, characterized in that, The optical reflection module includes a set of reflective mirrors respectively disposed on the inner side of the left and right A-pillars for reflecting optical images in the blind zone; The electronic display module includes a transparent display screen integrated on the surface of the A-pillar, used to receive real-time blind spot images acquired by a miniature wide-angle camera on the outside of the A-pillar, and to overlay the images with the blind spot optical images.

3. The system for eliminating the blind spot of the A-pillar of a vehicle according to claim 2, characterized in that, The optical reflection module includes: a first optical reflection group and a second optical reflection group symmetrically arranged on the inner side of the left and right A-pillars based on a sliding rail mounting structure; Each of the optical reflection groups includes: A reflector, wherein the reflector is an asymmetric curvature reflector, is used to acquire images of the blind spot of the opposite A-pillar; A presentation mirror, located below the reflector, is connected to the corresponding A-pillar via an adjustment device in the sliding rail mounting structure, and is used to present the light from the reflector to the driver's field of vision.

4. A system for eliminating blind spots of a vehicle's A-pillar according to claim 2, characterized in that, The electronic display module further includes: a microprism array that drives the transparent display screen; The microprism array is fabricated on the light-emitting layer of each sub-pixel of the transparent display screen. It is used to refract the display light reflected by the light-emitting layer to adjust the propagation direction of the display light. This allows the real-time blind spot image generated by the transparent display screen based on the display light to be spatially fused with the blind spot optical image to obtain composite visual content.

5. A system for eliminating blind spots of the A-pillar of a vehicle according to claim 4, characterized in that, The real-time blind zone image generated based on the displayed light is spatially fused with the blind zone optical image to obtain composite field-of-view content, specifically including: Based on the spatial pose relationship between the optical reflection module and the transparent display screen, establish the pixel coordinates of the real-time blind spot image and the mapping relationship between them and the projection optical path where the blind spot optical image is located; The real-time blind spot image is subjected to perspective correction according to the mapping relationship, so as to align the real-time blind spot image with the blind spot optical image; The aligned real-time blind spot image and the blind spot optical image are pixel-wise superimposed to generate the composite field of view content on the transparent display screen.

6. A system for eliminating blind spots caused by the A-pillar of a vehicle according to claim 1, characterized in that, Receiving multi-source sensing signals and dynamically adjusting the composite field of view content based on the multi-source sensing signals specifically includes: Receive multi-source perception signals to determine the current driving scenario and driver intention based on the multi-source perception signals, and match the vision adjustment priority based on the current driving scenario and driver intention to determine the vision compensation area and compensation data corresponding to the A-pillar blind spot of the vehicle. Based on the field of view compensation area and compensation data, an adjustment instruction is generated; The electronic display module is adjusted based on the adjustment command, and / or the optical path direction of the optical reflection module is adjusted to obtain the adjusted composite field of view content.

7. A system for eliminating blind spots of a vehicle's A-pillar according to claim 6, characterized in that, Receiving multi-source perception signals, determining the current driving scenario and driver intention based on the multi-source perception signals, and matching the vision adjustment priority based on the current driving scenario and driver intention to determine the vision compensation area and compensation data corresponding to the A-pillar blind spot of the vehicle, specifically including: Based on the multi-source perception data and preset thresholds, the current driving scenario and driver intent are determined; A preset priority mapping rule library is invoked to assign basic priority weights to the driving scenario model and the driver intention model. Based on the dynamic arbitration result of the current driving scenario and the driver intention, the basic priority weights are corrected to obtain the field of view adjustment priority. Based on the field of view adjustment priority, the field of view compensation area for the A-pillar blind spot on different priority sides is determined, and compensation data including electronic display enhancement parameters and optical reflection adjustment parameters is generated.

8. A system for eliminating blind spots of the A-pillar of a vehicle according to claim 1, characterized in that, The adjusted composite field of view is acquired, the risk characteristics of the vehicle's A-pillar blind spot are identified, and a graded safety response is performed based on the risk characteristics of the A-pillar blind spot, specifically including: Based on the adjusted composite field of view content, risk targets in the target detection area are identified; wherein, the target detection area is the projection area corresponding to the A-pillar blind spot; Feature extraction is performed on the risk target to obtain the risk characteristics of the vehicle's A-pillar blind spot; Based on the multi-source perception data, the current driving condition of the vehicle is determined. Based on the current driving condition and the risk characteristics, a risk assessment is performed on the risk target to obtain the risk level of the risk target. Trigger and execute graded security response operations corresponding to the risk level, continuously monitor the risk of the risk target, and adjust the composite view content and graded security response operations accordingly.

9. A system for eliminating blind spots caused by the A-pillar of a vehicle according to claim 1, characterized in that, The sensing module specifically includes: An eye-tracking module includes a camera located inside the vehicle, used to acquire images of the vehicle driver and determine the driver's head position and gaze direction based on the acquired images. The vehicle status perception module, integrated with the vehicle network, is used to acquire the vehicle's dynamic data via the vehicle CAN bus. An environmental perception module is used to acquire external environmental data of the vehicle.

10. A control method for eliminating a vehicle A-pillar blind spot system, wherein the A-pillar blind spot elimination system is any one of the vehicle A-pillar blind spot elimination systems according to claims 1-9; The control method for eliminating the blind spot of the vehicle's A-pillar specifically includes: The multimodal image fusion system overlays information from each module to generate composite visual content; wherein, the multimodal image fusion system is integrated into the left and right A-pillars of the vehicle and includes an optical reflection module and an electronic display module; The dynamic intelligent control system receives multi-source sensing signals from multiple sensing modules and dynamically adjusts the composite field of view content based on the multi-source sensing signals. Based on the adjusted composite field of view obtained by the active safety system, the risk characteristics of the vehicle's A-pillar blind spot are identified, and a graded safety response is performed based on the risk characteristics of the A-pillar blind spot.

Images