Vehicle external environment detection equipment based on V2X

Abstract

The invention belongs to the technical field of intelligent traffic and Internet of Vehicles, particularly relates to vehicle external environment detection equipment based on V2X, and aims to solve the problems that an existing scheme proposed in the background technology focuses on data sharing and lacks deep fusion and collaborative modeling capability of multi-source heterogeneous data. Comprising a V2X receiving and transmitting device supporting DSRC and C-V2X dual-mode communication protocols. By dynamically adjusting the weight of the sensor, the environment dependence of a single sensor is remarkably reduced, all-weather high-precision detection is realized, the problem of mismatching of multi-sensor data caused by clock skew or coordinate system difference is solved, a lane can be triggered in advance to keep assistance or the vehicle following distance can be adjusted, serial collision is effectively avoided, and the detection accuracy is improved. Potential cut-in risks can be predicted in advance, the response speed of an active safety system is remarkably improved, a path planning module generates a detouring track accordingly, the blind area defect of a traditional sensor is overcome, and a three-level early warning mechanism is triggered based on a fusion result of a D-S evidence theory and equipment.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation and vehicle networking technology, and in particular to a vehicle external environment detection device based on V2X. Background Art

[0002] Self-driving cars, also known as autonomous driving, unmanned riding or unmanned autonomous driving, use computer systems to control the flight of cars without the need for a driver. Before the development of autonomous driving technology, each part of the car was operated by a driver to complete different tasks, such as traffic light detection, vehicle control system coordination, braking, turning, etc. The principle of an autonomous car is that the computer system first inputs the current environmental information from various sensors, such as map information, road distance, vehicles in front of the vehicle, pedestrians, obstacles, etc., and then determines which action to take next through complex calculations. The next action can be directly fed back to the sensor, such as changing the direction of the vehicle or moving forward. Autonomous driving technology mainly uses computers to synthesize complex driving environments to achieve real-time control of the vehicle, and combines sensors, GPS and map information, as well as some specific algorithms to send instructions to various parts of the car in turn to achieve the purpose of coordinated control.

[0003] Existing autonomous vehicles mainly rely on on-board sensors (such as cameras, radars, and lidars) for environmental perception, which has the following defects:

[0004] 1. The sensor is easily affected by weather, occlusion and other factors, and the detection distance and accuracy are limited;

[0005] 2. Unable to obtain dynamic information in the Non-Line-of-Sight (NLOS) area;

[0006] 3. The coverage of a single vehicle sensor is limited and it is difficult to cope with complex traffic scenarios.

[0007] Therefore, although V2X technology can realize information interaction, existing solutions mostly focus on data sharing and lack the ability to deeply integrate and collaboratively model multi-source heterogeneous data. Summary of the invention

[0008] In view of the shortcomings of the prior art, the present invention provides a vehicle external environment detection device based on V2X, which overcomes the shortcomings of the prior art and effectively solves the problem that the existing solutions focus more on data sharing and lack the ability to deeply integrate and collaboratively model multi-source heterogeneous data.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] A vehicle external environment detection device based on V2X, including: supporting DSRC and

[0011] V2X transceiver for C-V2X dual-mode communication protocol;

[0012] Heterogeneous data interfaces connected to vehicle-mounted sensors, including cameras, millimeter-wave radars, and lidars;

[0013] Collaborative calibration module for spatiotemporal alignment of multi-source data Multi-source data fusion processor based on Bayesian reasoning;

[0014] A control unit for generating a dynamic environment model and outputting decision instructions, wherein the environment model is output by the control unit;

[0015] NLOS compensation model for edge computing unit and device integration for real-time processing of camera image data.

[0016] Preferably, the collaborative calibration module includes a dual redundant global clock synchronization mechanism based on the NTP protocol and the PTP protocol, and a dynamic coordinate system conversion algorithm based on the roadside unit (RSU), wherein the dynamic coordinate system conversion algorithm supports real-time mapping of the WGS84 coordinate system and the local coordinate system.

[0017] Preferably, the multi-source data fusion processor includes a channel quality assessment submodule and an adaptive filter, wherein the channel quality assessment submodule is used to analyze the signal-to-noise ratio, packet loss rate and transmission delay of the V2X data packet, and the adaptive filter dynamically adjusts the fusion weight according to the data freshness, signal strength and channel quality.

[0018] Preferably, the adaptive filter adopts a hybrid model of Kalman filtering and particle filtering, and introduces a confidence threshold to filter low-reliability data.

[0019] Preferably, the environmental model includes a three-dimensional occupancy grid map updated in real time, trajectory prediction data of traffic participants, and emergency event semantic information based on V2X broadcast, wherein the three-dimensional occupancy grid map is used to mark the location and confidence level of obstacles, the trajectory prediction data of traffic participants includes velocity vectors and motion intention analysis results, and the emergency event semantic information includes traffic accidents, road construction and sudden weather.

[0020] Preferably, the edge computing unit deploys a lightweight convolutional neural network (CNN) model and performs feature-level fusion with the V2X data.

[0021] Preferably, the NLOS compensation model obtains the reflection signal of the blocked area through RSU relay, and reconstructs the virtual perception field based on the geometric topological relationship.

[0022] Preferably, the NLOS compensation model adopts a multipath signal analysis algorithm and dynamically optimizes compensation accuracy in combination with RSU deployment density.

[0023] Preferably, the vehicle external environment detection method of the device comprises the following steps:

[0024] Step S1: receiving and parsing BSM (Basic Safety Message), SPAT (Signal Phase and Timing) and MAP (Map Data) in V2X messages;

[0025] Step S2: Perform spatiotemporal calibration on multi-source data to eliminate clock bias and coordinate system differences;

[0026] Step S3: Execute multimodal data fusion based on DS evidence theory to generate environmental perception results;

[0027] Step S4: triggering graded warnings based on the fusion results, including HUD prompts, sound alarms, and automatic emergency braking.

[0028] Preferably, the spatiotemporal calibration step includes using the high-precision clock signal broadcast by the RSU to synchronize devices and aligning the coordinate system of the local environment in real time through SLAM (Synchronous Localization and Mapping) technology.

[0029] The beneficial effects of the present invention are:

[0030] 1. The V2X-based vehicle external environment detection device of the present invention has full scene coverage and high-precision perception. By supporting DSRC and C-V2X dual-mode communication protocols, combined with the heterogeneous fusion of vehicle-mounted sensors (cameras, millimeter-wave radars, lidars) and V2X data, the device can simultaneously obtain dynamic information in line-of-sight (LOS) and non-line-of-sight (NLOS) areas. For example, in an intersection scenario, the reflected signal relayed by the RSU can reconstruct the position of obscured pedestrians with a confidence level of 90%; in severe weather (such as fog and heavy rain), by dynamically adjusting the sensor weight (such as the V2X data weight is increased to 0.8), the environmental dependence of a single sensor is significantly reduced, achieving all-weather high-precision detection.

[0031] 2. The V2X-based vehicle external environment detection device of the present invention has deep fusion of multi-source data and guarantees of spatiotemporal consistency. The device adopts a multi-source data fusion processor based on Bayesian reasoning, combined with dual redundant global clock synchronization (NTP / PTP protocol) and dynamic coordinate system conversion algorithm (real-time mapping of WGS84 and local coordinate system), which solves the mismatch problem caused by clock deviation or coordinate system difference in multi-sensor data. For example, in the urban intersection scene, after the laser radar point cloud and V2X data are aligned in time and space, the error is controlled at the centimeter level, providing a reliable data basis for trajectory prediction and collision risk assessment.

[0032] 3. The V2X-based vehicle external environment detection equipment of the present invention supports dynamic environment modeling and intelligent decision optimization. The three-dimensional occupancy grid map generated by the control unit supports real-time update, annotates the obstacle position and confidence level, and combines the trajectory prediction data of traffic participants (speed vector, motion intention analysis) and the emergency semantic information (traffic accidents, road construction, etc.) broadcast by V2X to achieve dynamic environment modeling. For example, in the highway scene, the device obtains the accident information 1 km ahead through V2V communication, triggers lane keeping assistance or adjusts the following distance in advance, and effectively avoids chain collisions.

[0033] 4. The V2X-based vehicle external environment detection device of the present invention, edge computing and low-latency response edge computing, unit deployment of lightweight CNN model, real-time processing of camera image data (such as identifying vehicle lane change intention), and feature-level fusion with V2X data. In highway scenarios, this technology can predict potential cut-in risks 1.5 seconds in advance, significantly improving the response speed of the active safety system.

[0034] 5. The vehicle external environment detection device based on V2X of the present invention, NLOS compensation and virtual perception field expansion, the NLOS compensation model reconstructs the virtual perception field of the blocked area through the multipath signal analysis algorithm and RSU relay. For example, in a dense fog environment, the construction area information transmitted by the RSU can be used to mark the location of the cone barrel on the map, and the path planning module generates a detour trajectory based on this to make up for the blind spot defects of traditional sensors.

[0035] 6. The vehicle external environment detection device based on V2X of the present invention triggers a three-level warning mechanism based on the fusion result of DS evidence theory: HUD warning, sound alarm and automatic emergency braking (AEB). For example, when the predicted collision time is less than 2 seconds, AEB intervenes and achieves 6m / s 2 deceleration to minimize accident damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a general flow chart of a vehicle external environment detection device based on V2X proposed by the present invention;

[0037] Figure 2 A flowchart of a collaborative calibration module of a vehicle external environment detection device based on V2X proposed by the present invention;

[0038] Figure 3 A multi-source data fusion flow chart of a vehicle external environment detection device based on V2X proposed by the present invention;

[0039] Figure 4 A flow chart of an NLOS compensation model of a vehicle external environment detection device based on V2X proposed by the present invention;

[0040] Figure 5 This is a flow chart of a detection method for a vehicle external environment detection device based on V2X proposed in the present invention. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0042] Reference Figure 1-Figure 5 , a vehicle external environment detection device based on V2X, comprising:

[0043] V2X transceiver supporting DSRC and C-V2X dual-mode communication protocols;

[0044] Heterogeneous data interfaces connected to vehicle-mounted sensors, including cameras, millimeter-wave radars, and lidars;

[0045] Collaborative calibration module for spatiotemporal alignment of multi-source data Multi-source data fusion processor based on Bayesian reasoning;

[0046] A control unit for generating a dynamic environment model and outputting decision instructions, wherein the environment model is output by the control unit;

[0047] NLOS compensation model for edge computing unit and device integration for real-time processing of camera image data.

[0048] The collaborative calibration module includes a dual redundant global clock synchronization mechanism based on the NTP protocol and the PTP protocol, and a dynamic coordinate system conversion algorithm based on the roadside unit (RSU), wherein the dynamic coordinate system conversion algorithm supports real-time mapping between the WGS84 coordinate system and the local coordinate system.

[0049] The multi-source data fusion processor includes a channel quality assessment submodule and an adaptive filter, wherein the channel quality assessment submodule is used to analyze the signal-to-noise ratio, packet loss rate and transmission delay of the V2X data packet, and the adaptive filter dynamically adjusts the fusion weight according to the data freshness, signal strength and channel quality.

[0050] The adaptive filter adopts a hybrid model of Kalman filtering and particle filtering, and introduces a confidence threshold to filter low-reliability data.

[0051] The environmental model includes a three-dimensional occupancy grid map updated in real time, trajectory prediction data of traffic participants, and emergency event semantic information based on V2X broadcast, wherein the three-dimensional occupancy grid map is used to mark the location and confidence level of obstacles, the trajectory prediction data of traffic participants includes velocity vectors and motion intention analysis results, and the emergency event semantic information includes traffic accidents, road construction and sudden weather.

[0052] The edge computing unit deploys a lightweight convolutional neural network (CNN) model and performs feature-level fusion with the V2X data.

[0053] The NLOS compensation model obtains the reflected signal of the blocked area through the RSU relay and reconstructs the virtual perception field based on the geometric topological relationship.

[0054] The NLOS compensation model adopts a multipath signal analysis algorithm and combines the RSU deployment density to dynamically optimize the compensation accuracy.

[0055] The vehicle external environment detection method of the device comprises the following steps:

[0056] Step S1: receiving and parsing BSM (Basic Safety Message), SPAT (Signal Phase and Timing) and MAP (Map Data) in V2X messages;

[0057] Step S2: Perform spatiotemporal calibration on multi-source data to eliminate clock bias and coordinate system differences;

[0058] Step S3: Execute multimodal data fusion based on DS evidence theory to generate environmental perception results;

[0059] Step S4: triggering graded warnings based on the fusion results, including HUD prompts, sound alarms, and automatic emergency braking.

[0060] The spatiotemporal calibration step includes using the high-precision clock signal broadcast by the RSU to synchronize devices and aligning the local environment's coordinate system in real time through SLAM (Simultaneous Localization and Mapping) technology.

[0061] Example 1: Urban intersection scenario

[0062] Step 1: Data Collection

[0063] The main vehicle receives the BSM data (including GPS position, speed, and heading angle) of neighboring vehicles through the V2X transceiver, and also receives the SPAT signal light countdown information sent by the roadside unit (RSU) and the V2P (Vehicle-to-Pedestrian) signal of pedestrians in the blind spot.

[0064] On-board sensors (lidar, millimeter-wave radar, and cameras) simultaneously collect local environmental data, including obstacle distances, pedestrian outlines, and traffic light status.

[0065] Step 2: Space-time alignment

[0066] The collaborative calibration module uses the RSU as the benchmark, synchronizes the clocks between devices to microsecond accuracy through the NTP protocol, and uses a dynamic coordinate system conversion algorithm to unify the V2X data and the lidar point cloud into the WGS84 coordinate system (error <5cm).

[0067] SLAM technology builds a local environment map in real time and aligns it with the global coordinate system to eliminate positioning deviations (such as Figure 2 shown).

[0068] Step 3: Multi-source data fusion

[0069] The multi-source data fusion processor dynamically adjusts the weights based on the channel quality assessment results (signal-to-noise ratio > 20dB, packet loss rate <1%): in rainy and foggy weather, the weight of millimeter-wave radar data is set to 0.7 and the weight of V2X data is set to 0.3.

[0070] The adaptive filter adopts a hybrid model of Kalman filtering and particle filtering to filter low reliability data (such as noise interference point cloud) with a confidence level lower than 80%.

[0071] Step 4: NLOS compensation and virtual perception

[0072] The NLOS compensation model obtains the reflected signal of the bus-blocked area through the RSU relay, reconstructs the pedestrian position (confidence 90%) based on the geometric topological relationship, and inputs the data into the three-dimensional occupancy grid map (such as Figure 3 shown).

[0073] Step 5: Environmental modeling and decision execution

[0074] The control unit generates a real-time updated three-dimensional map, annotates the status of pedestrians, vehicles and traffic lights, and calculates the risk of pedestrian crossing based on a trajectory prediction algorithm (collision probability > 95%).

[0075] When the predicted collision time is less than 2 seconds, a level 3 warning is triggered:

[0076] Level 1: The HUD displays a red warning box to prompt the driver to perform emergency braking;

[0077] Level 2: The buzzer emits an urgent alarm sound (frequency 2kHz);

[0078] Level 3: Automatic emergency braking (AEB) intervention, achieving 6m / s 2 The deceleration rate (such as Figure 4 shown).

[0079] Example 2: Highway scenario

[0080] Step 1: Long-distance information acquisition

[0081] V2X communication receives broadcast information about traffic accidents 1 km ahead (including accident coordinates, number of vehicles and lane occupancy status), and the collaborative calibration module maps the accident point to a local map (with an accuracy of ±1m).

[0082] Step 2: Multi-vehicle collaborative perception

[0083] Acquire the emergency braking status of vehicles in adjacent lanes (deceleration > 4m / s) through V2V communication 2 ), after integrating the vehicle's millimeter-wave radar data, adjust the following distance to the safety threshold (minimum 50m).

[0084] Step 3: Edge computing and risk prediction

[0085] The lightweight CNN model deployed in the edge computing unit identifies the lane-changing intention of the vehicle ahead (such as activation of the turn signal and wheel deflection angle > 5°) in real time, and predicts potential cut-in risks in combination with V2X data.

[0086] When the risk level reaches the threshold, the lane keeping assist (LKA) is triggered 1.5 seconds in advance, and the steering wheel torque output is 3N·m (such as Figure 5 shown).

[0087] Example 3: Severe Weather Scenario

[0088] Step 1: Dynamic adjustment of sensor weights

[0089] In a foggy environment, the camera fails due to insufficient visibility. The control unit increases the V2X data weight to 0.8 and reduces the millimeter-wave radar weight to 0.2 to ensure perception continuity.

[0090] Step 2: Virtual Perception Field Expansion

[0091] The RSU relays the information of the construction area obscured by the dense fog (such as the coordinates of the cone barrel and the construction scope), the control unit marks the location of the obstacle on the map, and the path planning module generates a detour trajectory (lateral offset of 2.5m).

[0092] Step 3: Optimize the hierarchical warning

[0093] Due to low visibility, the system adjusts the color of the HUD warning frame to high-contrast yellow and reduces the buzzer frequency to 1kHz to reduce interference.

[0094] Example 4: Tunnel environment scene

[0095] Step 1: GPS signal compensation

[0096] When the GPS signal is lost in the tunnel, the device switches to SLAM technology combined with V2X data (such as the internal map of the tunnel provided by the RSU) to achieve precise positioning (error <0.3m).

[0097] Step 2: Multimodal Data Redundancy

[0098] The laser radar and millimeter wave radar data are fused to construct a three-dimensional point cloud map of the tunnel, marking static obstacles (such as maintenance vehicles) and dynamic targets (such as motorcycles).

[0099] Step 3: Emergency Braking Coordination

[0100] If the vehicle ahead brakes suddenly (deceleration> 8m / s 2 ), the system broadcasts warning information through V2V, and the rear vehicle triggers AEB synchronously to form a collaborative braking chain.

[0101] Example 5: Nighttime driving scenario

[0102] Step 1: Low-light enhancement

[0103] The camera switches to infrared mode, and the edge computing unit uses the CNN model to enhance image contrast and identify reflective signs of pedestrians and animal outlines.

[0104] Step 2: V2X complements sensors

[0105] The RSU provides nighttime road construction information (such as temporary diversion signs), which is then integrated with the on-board radar data to optimize route planning (such as adjusting the speed limit to 30 km / h).

[0106] Step 3: Anti-glare treatment

[0107] The control unit dynamically adjusts the HUD brightness and warning frame position according to the light intensity of oncoming vehicles (>2000 lumens) to avoid visual interference to the driver.

[0108] Equipment structure and working principle:

[0109] The core modules of the device include:

[0110] 1.V2X transceiver: supports DSRC (5.9GHz frequency band) and C-V2X (LTE / 5G) dual-mode communications, with a maximum coverage radius of 1.5km.

[0111] 2. Collaborative calibration module: Integrates dual redundant clock synchronization mechanism (NTP error <1μs, PTP error <100ns) and SLAM algorithm to ensure temporal and spatial consistency.

[0112] 3. Multi-source data fusion processor: adopts hybrid filtering model (Kalman filtering + particle filtering), and the data update frequency is 100Hz.

[0113] 4.NLOS compensation model: Based on multipath signal analysis (delay spread <50ns), the compensation accuracy reaches 85%.

[0114] 5. Edge computing unit: equipped with a lightweight CNN model (parameter quantity <1M), processing delay <10ms.

[0115] The equipment workflow is:

[0116] 1. Data collection → 2. Space-time alignment → 3. Data fusion → 4. Environmental modeling → 5. Decision execution, full closed-loop control to ensure real-time and reliability.

[0117] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0118] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0119] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A vehicle external environment detection device based on V2X, characterized in that: include: V2X transceiver supporting DSRC and C-V2X dual-mode communication protocols; Heterogeneous data interfaces connected to vehicle-mounted sensors, including cameras, millimeter-wave radars, and lidars; Collaborative calibration module for spatiotemporal alignment of multi-source data Multi-source data fusion processor based on Bayesian reasoning; A control unit for generating a dynamic environment model and outputting decision instructions, wherein the environment model is output by the control unit; NLOS compensation model for edge computing unit and device integration for real-time processing of camera image data.

2. The vehicle external environment detection device based on V2X according to claim 1, characterized in that: The collaborative calibration module includes a dual redundant global clock synchronization mechanism based on the NTP protocol and the PTP protocol, and a dynamic coordinate system conversion algorithm based on the roadside unit (RSU), wherein the dynamic coordinate system conversion algorithm supports real-time mapping between the WGS84 coordinate system and the local coordinate system.

3. The vehicle external environment detection device based on V2X according to claim 1, characterized in that: The multi-source data fusion processor includes a channel quality assessment submodule and an adaptive filter, wherein the channel quality assessment submodule is used to analyze the signal-to-noise ratio, packet loss rate and transmission delay of the V2X data packet, and the adaptive filter dynamically adjusts the fusion weight according to the data freshness, signal strength and channel quality.

4. The vehicle external environment detection device based on V2X according to claim 3 is characterized in that: The adaptive filter adopts a hybrid model of Kalman filtering and particle filtering, and introduces a confidence threshold to filter low-reliability data.

5. The vehicle external environment detection device based on V2X according to claim 1, characterized in that: The environmental model includes a three-dimensional occupancy grid map updated in real time, trajectory prediction data of traffic participants, and emergency event semantic information based on V2X broadcast, wherein the three-dimensional occupancy grid map is used to mark the location and confidence level of obstacles, the trajectory prediction data of traffic participants includes velocity vectors and motion intention analysis results, and the emergency event semantic information includes traffic accidents, road construction and sudden weather.

6. The vehicle external environment detection device based on V2X according to claim 1, characterized in that: The edge computing unit deploys a lightweight convolutional neural network (CNN) model and performs feature-level fusion with the V2X data.

7. The vehicle external environment detection device based on V2X according to claim 1, characterized in that: The NLOS compensation model obtains the reflected signal of the blocked area through the RSU relay and reconstructs the virtual perception field based on the geometric topological relationship.

8. The vehicle external environment detection device based on V2X according to claim 7, characterized in that: The NLOS compensation model adopts a multipath signal analysis algorithm and dynamically optimizes compensation accuracy in combination with RSU deployment density.

9. A vehicle external environment detection method based on the device according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step S1: receiving and parsing BSM (Basic Safety Message), SPAT (Signal Phase and Timing) and MAP (Map Data) in V2X messages; Step S2: Perform spatiotemporal calibration on multi-source data to eliminate clock bias and coordinate system differences; Step S3: Execute multimodal data fusion based on DS evidence theory to generate environmental perception results; Step S4: triggering graded warnings based on the fusion results, including HUD prompts, sound alarms, and automatic emergency braking.

10. The method according to claim 9, characterized in that The spatiotemporal calibration step includes using the high-precision clock signal broadcast by the RSU to synchronize devices and aligning the local environment's coordinate system in real time through SLAM (Simultaneous Localization and Mapping) technology.

Images