Vehicle early warning control method, system and vehicle

By obtaining TSR identification information and vehicle-road perception fusion information, judging and activating vehicle early warning scenarios, the problem of not being able to obtain real and effective vehicle driving environment information in the prior art is solved, accurate early warning signal transmission is achieved, and the vehicle's safe driving ability is improved .

CN115482685BActive Publication Date: 2024-11-22BEIJING NEW ENERGY VEHICLE TECH INNOVATION CENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210905252.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-29
Publication Date
2024-11-22
Estimated Expiration
2042-07-29

AI Technical Summary

Technical Problem

The prior art cannot obtain real and effective vehicle driving environment information in a timely manner, resulting in the inability to accurately transmit early warning signals, resulting in vehicle accidents.

Method used

By obtaining TSR identification information, we judge the scene of activating the vehicle warning function, and combining vehicle-road perception fusion information, we judge whether early warning instructions need to be issued to the driver, including the status judgment of traffic lights and traffic signs, to ensure that the driver receives accurate early warnings information.

Benefits of technology

It realizes that vehicles obtain traffic information in advance, judge and issue early warning instructions to ensure safe driving and reduce the occurrence of traffic accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115482685B_ABST
    Figure CN115482685B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of vehicle early warning, and discloses a vehicle early warning control method, system and vehicle, including the following steps: S1: obtain TSR identification information, and determine whether the TSR identification information belongs to the activation scene of activating the vehicle early warning function, if so, execute S2, otherwise terminate the process; S2: after the vehicle early warning function is activated, the vehicle enters the perception area to perform vehicle-road perception fusion to obtain vehicle-road perception fusion information; S3: determine whether it is necessary to issue an early warning instruction to the driver through the vehicle-road perception fusion information, if necessary, control the vehicle to issue an early warning instruction to the driver, if not, terminate the process. By pre-storing the activation scene on the vehicle early warning control system, and then determining whether the vehicle activates the relevant scene flag through the obtained TSR identification information, after determining the activation scene flag, further obtain more detailed traffic information status ahead through the vehicle-road perception fusion information, so that the vehicle can obtain the information status of the traffic light or traffic sign ahead in advance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle early warning, and in particular to a vehicle early warning control method, system and vehicle. Background Art

[0002] Safe driving is the first rigid demand of car users. During driving, vehicle collision is the main factor causing traffic accidents. For example, when passing through an intersection, the main cause of vehicle collision is the large vehicle blocking the traffic light or traffic sign, or the traffic light or traffic sign is blurred, or the vehicle cannot accurately obtain the information of the traffic light or traffic sign due to limited perception ability.

[0003] The existing technology obtains traffic information at the intersection ahead in advance through the TSR information broadcast function on the vehicle. However, since the actual existing information at the intersection cannot be transmitted to the vehicle in time and integrated with the vehicle-side information, it is impossible to obtain real and effective vehicle driving environment information, and thus it is impossible to transmit accurate vehicle driving warning signals to the driver. Summary of the invention

[0004] In order to solve the problem in the prior art that vehicle accidents occur due to the inability to obtain true and effective vehicle driving environment information and transmit accurate warning signals, the present invention provides a vehicle warning control method, system and vehicle.

[0005] The technical contents of the present invention include:

[0006] A vehicle early warning control method comprises the following steps:

[0007] S1: Obtain TSR identification information and determine whether the TSR identification information belongs to the activation scenario for activating the vehicle warning function. If so, execute S2, otherwise terminate the process;

[0008] S2: After the vehicle warning function is activated, the vehicle enters the perception area to perform vehicle-road perception fusion to obtain vehicle-road perception fusion information;

[0009] S3: Determine whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information. If necessary, control the vehicle to issue a warning instruction to the driver. If not, terminate the process.

[0010] Further, the S1 includes:

[0011] The activation scenarios include:

[0012] Traffic lights are blocked during TSR recognition;

[0013] Traffic lights are blurred or easily confused when TSR is used for recognition;

[0014] TSR recognizes scenes where traffic signs are blurry or easily confused.

[0015] Further, the S2 includes:

[0016] The vehicle-road perception fusion information includes a combination of vehicle-side perception fusion information and roadside perception fusion information.

[0017] Further, the S3 includes: when judging whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information, obtaining a judgment result;

[0018] The judgment results include:

[0019] When the traffic light ahead is green, calculate whether the vehicle can safely pass the intersection safety line based on the current vehicle motion state;

[0020] When the traffic light ahead is red or yellow, the vehicle warning distance is calculated based on the current vehicle motion status;

[0021] When the traffic sign ahead indicates a speed limit or other specific area where speed reduction is required, the current vehicle motion state is combined to calculate whether the vehicle exceeds the maximum speed limit.

[0022] Furthermore, when the state of the traffic light ahead is green, calculating whether the vehicle can safely pass the intersection safety line in combination with the current vehicle motion state includes:

[0023] If the vehicle cannot safely pass the intersection safety line, the control vehicle will issue a warning instruction to the driver;

[0024] If the vehicle can safely pass the intersection safety line, the process is terminated and the driver drives normally.

[0025] Furthermore, when the traffic sign information ahead indicates a speed limit or other specific area where the speed reduction is required, the calculation of whether the vehicle exceeds the maximum speed limit is performed in combination with the current vehicle motion state, including:

[0026] If the vehicle does not exceed the maximum speed limit, the control warning system will issue normal driving instructions to the driver;

[0027] If the vehicle exceeds the maximum speed limit, the warning system will be controlled to issue a driving warning instruction to the driver.

[0028] The present invention also provides a vehicle early warning control system, which implements the steps of any of the above-mentioned vehicle early warning control methods, including:

[0029] An acquisition unit, used for acquiring TSR identification information;

[0030] A first judgment unit, used to judge whether the TSR identification information belongs to an activation scenario for activating a vehicle warning function;

[0031] A perception unit, used to obtain vehicle-road perception fusion information;

[0032] The second judgment unit is used to judge whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information;

[0033] The early warning unit is used to control the vehicle to issue early warning instructions to the driver.

[0034] Furthermore, the sensing unit includes:

[0035] Roadside unit, used to obtain roadside end fusion perception information;

[0036] The vehicle-mounted unit is used to obtain vehicle-side fusion perception information.

[0037] The present invention also provides a vehicle for implementing the steps of any of the above-mentioned vehicle early warning control methods.

[0038] The beneficial effects of the present invention include at least: the present invention pre-stores the activation scene in the vehicle warning control system, and then determines whether the vehicle activates the relevant scene flag through the obtained TSR identification information. After determining that the scene flag is activated, the vehicle-road perception fusion information is further used to obtain more detailed traffic information status ahead, so that the vehicle can obtain the information status of the traffic lights or traffic signs ahead in advance, perceive and judge the traffic environment in advance, and determine whether it is necessary to issue warning instructions to the driver based on the judgment, such as alarm, active deceleration or emergency braking and other related warning instructions, so that the vehicle can achieve the purpose of safe driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of the flow structure of a vehicle early warning control method provided in Example 1 of the present invention.

[0040] Figure 2 Schematic diagram of the process architecture of the roadside perception fusion algorithm provided in Example 1 of the present invention.

[0041] Figure 3 Schematic diagram of the process architecture of the vehicle-road perception fusion algorithm provided in Example 1 of the present invention.

[0042] Figure 4 This is a schematic diagram of the flow structure of a vehicle early warning control system provided in Example 2 of the present invention.

[0043] Figure 5 This is a schematic diagram of the flow structure of a vehicle early warning implementation method in Example 3 of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Example 1

[0046] Combination Figure 1-3 As shown, the present invention provides a vehicle early warning control method, comprising the following steps:

[0047] S1: Obtain TSR identification information and determine whether the TSR identification information belongs to the activation scenario for activating the vehicle warning function. If so, execute S2, otherwise terminate the process;

[0048] S2: After the vehicle warning function is activated, the vehicle enters the perception area to perform vehicle-road perception fusion to obtain vehicle-road perception fusion information;

[0049] S3: Determine whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information. If necessary, control the vehicle to issue a warning instruction to the driver. If not, terminate the process.

[0050] Further, the S1 includes:

[0051] The activation scenarios include:

[0052] Traffic lights are blocked during TSR recognition;

[0053] Traffic lights are blurred or easily confused when TSR is used for recognition;

[0054] TSR recognizes scenes where traffic signs are blurry or easily confused.

[0055] Further, the S2 includes:

[0056] The vehicle-road perception fusion information is a combination of vehicle-side perception fusion information and roadside perception fusion information.

[0057] Further, the S3 includes: when judging whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information, obtaining a judgment result;

[0058] The judgment results include:

[0059] When the traffic light ahead is green, calculate whether the vehicle can safely pass the intersection safety line based on the current vehicle motion state;

[0060] When the traffic light ahead is red or yellow, the vehicle warning distance is calculated based on the current vehicle motion status;

[0061] When the traffic sign ahead indicates a speed limit or other specific area where speed reduction is required, the current vehicle motion state is combined to calculate whether the vehicle exceeds the maximum speed limit.

[0062] Furthermore, when the state of the traffic light ahead is green, calculating whether the vehicle can safely pass the intersection safety line in combination with the current vehicle motion state includes:

[0063] If the vehicle cannot safely pass the intersection safety line, the control vehicle will issue a warning instruction to the driver;

[0064] If the vehicle can safely pass the intersection safety line, the process is terminated and the driver drives normally.

[0065] Furthermore, when the traffic sign information ahead indicates a speed limit or other specific area where the speed reduction is required, the calculation of whether the vehicle exceeds the maximum speed limit is performed in combination with the current vehicle motion state, including:

[0066] If the vehicle does not exceed the maximum speed limit, the control warning system will issue normal driving instructions to the driver;

[0067] If the vehicle exceeds the maximum speed limit, the warning system will be controlled to issue a driving warning instruction to the driver.

[0068] In this embodiment, traffic light status information or traffic sign information is obtained by a roadside unit and sent to the vehicle-mounted unit. The communication coverage range of the roadside unit can reach 200-300 meters. Therefore, the warning distance is mainly considered when the vehicle passes through a traffic light intersection or traffic sign in compliance with traffic regulations, and when a driving warning information is issued when there is a driving risk.

[0069] The activation scenarios of the vehicle warning control method include three situations. When the traffic lights at the intersection are blocked during TSR recognition, or when the traffic lights at the intersection are blurred or easily confused during TSR recognition, whether the vehicle should be warned mainly considers two aspects: one is the traffic light status information, and the other is the warning distance.

[0070] First, after determining the direction of travel of the vehicle's current lane, the traffic light status is determined through V2X information. When the traffic light is in red / yellow status, a comprehensive judgment is made on the vehicle's motion status and traffic light information to obtain the vehicle's current warning distance. When the vehicle enters the warning distance, a reminder is issued to the driver. After receiving the reminder, the driver actively takes braking measures and implements deceleration measures.

[0071] Calculate the distance that the vehicle can travel to stop at the stop line with a small deceleration under ideal conditions, that is, the warning distance of the vehicle, and calculate it through the following formula:

[0072]

[0073] Among them, v SV is the initial vehicle speed, t svd ,t rbr are the driver reaction time and the braking system response time, respectively; a1 is the mild braking deceleration, which is a vector value, acceleration is a positive value and deceleration is a negative value; d pre is the warning safety distance. Since the vehicle slows down to a stop (speed is 0km / h), -v in the formula 2 sv is 0 2 -v 2 sv Simplified.

[0074] When taking v SV =72km / h, t svd =0.5s, t rbr =0.6s, a1=0.2g, b pre =25m, it can be calculated that S0=149m at this time. That is, after the vehicle enters the communication range of the roadside unit, after a comprehensive judgment of the vehicle's motion state and traffic light information, a warning instruction is issued to the driver after the vehicle enters the warning distance S0. After receiving the warning instruction, the driver takes the initiative to take braking measures to decelerate.

[0075] When the traffic light is in the green state, it is necessary to further determine whether the green light is in the countdown state. If the green light has started counting down, it is necessary to further determine whether the vehicle can safely pass through the intersection. If it cannot pass, it is necessary to stop safely outside the vehicle stop line and issue a timely warning, otherwise it can pass normally.

[0076] When the vehicle cannot pass through the intersection before the green light ends, a judgment is made based on the vehicle's driving information, and formula (2) is used to calculate whether the vehicle can pass through the intersection, that is:

[0077]

[0078] If TTC>t sv , the vehicle can pass normally; if TTC<t sv , the vehicle issues a warning, and the warning distance is calculated according to formula (1). sv is the distance from the main vehicle to the safety line, v sv is the current speed of the main vehicle, t sv The time it takes for the main vehicle to reach the safety line at the current speed.

[0079] When the last activation scenario occurs: the traffic sign is blurred or easily confused during TSR recognition.

[0080] When encountering such a situation, whether it is necessary to issue a warning to the driver is mainly determined by two aspects: one is the traffic sign information, and the other is the warning distance. First, the traffic sign is measured and located through the roadside unit, and then the traffic sign information and location are entered into the roadside unit and sent to the vehicle-mounted unit.

[0081] Traffic sign information mainly includes speed limit signs, construction signs, school zone signs, etc. At this time, the vehicle speed v sv The vehicle is then given a warning based on the speed limit information. If there is no additional speed limit sign in the school area, the vehicle will pass through the road according to the traffic regulations at a speed limit of 30 km / h. If the vehicle speed is greater than the speed limit information, the vehicle will be decelerated to the maximum speed limit to pass the road section. Otherwise, the vehicle will drive normally. The distance that the vehicle decelerates to the maximum speed limit is calculated by formula (3):

[0082]

[0083] Among them, S lim is the distance required for the vehicle to decelerate to the maximum speed limit, v lim is the maximum speed limit on the traffic sign, t svd ,t rbr are the driver reaction time and the braking system response time respectively. a1 is the mild braking deceleration, which is a vector value. Acceleration is a positive value and deceleration is a negative value.

[0084] When taking v lim =72km / h, t svd =0.5s, t rbr =0.6s, a1=0.2g, v lim =30km / h, we can calculate S lim =131m.

[0085] That is, after the vehicle enters the communication range of the roadside unit, the vehicle movement state and traffic sign information are comprehensively judged. lim After that, a warning instruction is issued to the driver. After receiving the vehicle warning instruction, the driver takes the initiative to take braking measures to decelerate to the specified speed and pass safely.

[0086] The vehicle makes a comprehensive judgment on the vehicle through the vehicle perception fusion information, and then decides whether the vehicle needs to issue a warning instruction based on the judgment result. When the vehicle sends the traffic light information or traffic sign information obtained by the roadside unit to the on-board unit, it combines it with the vehicle motion state and vehicle motion environment obtained by the on-board unit to obtain more accurate traffic information. That is, the dangerous targets perceived by the vehicle side are obtained through vehicle-side fusion, and the roadside unit fuses the targets perceived by the roadside and broadcasts them to the vehicle. The perceived targets on the vehicle side are synchronized with the received roadside perceived targets in time and space, and finally outputs the most dangerous target based on the vehicle-road fusion perception to the warning unit to determine whether a warning is needed.

[0087] Specifically, the roadside unit will integrate the perceived traffic light information or traffic sign information, and then the on-board unit will integrate the perceived vehicle motion status and vehicle motion environment. Finally, the roadside fusion information of the roadside unit is integrated with the vehicle-side fusion information of the on-board unit to obtain more accurate traffic information for the vehicle to determine whether an early warning is needed.

[0088] The target after vehicle-side fusion processing mainly contains information including timestamp, target ID, coordinates (the distance between the target and the center line of the vehicle), relative speed V x 、V y , acceleration A x Etc. When the vehicle controller receives the target perception result from the vehicle, the spatial position of the camera and radar is first calibrated, and then the time synchronization method based on Lagrange interpolation is used to synchronize the camera and radar data; after the camera and radar data are synchronized in time and space, it is necessary to pair the targets from different sensors according to certain rules. In this embodiment, the global nearest neighbor (GNN) algorithm is used to match the targets. When the performance function E of the GNN algorithm is within the threshold range of 0 to Eth and the E value is the smallest, the two are considered to be the same target. Then the Kalman filter method is used to track the data of the matched targets and the targets that are not matched by the camera / radar, and the weighted method is used to fuse the matched target data. Finally, the designed rule method is used to select the CIPV / P targets, the targets that are not matched by the camera, the CIPV / P targets converted to absolute positions, and the parameters of the targets that are not matched by the camera converted to absolute positions as the input of the perception unit and the early warning unit.

[0089] The vehicle-side fusion process includes:

[0090] 1. Preprocess different information perceived by different sensors, including processing the X, Y, Vx, Vy, Ax, and Ay of the targets measured by the sensors, and removing invalid targets, facing targets, and empty targets;

[0091] 2. Convert the target data collected by different sensors to the same coordinate system for spatial synchronization, and then use the spatiotemporal synchronization algorithm, namely the Lagrange interpolation method, to synchronize the target data X, Y, Vx, Vy, Ax, and Ay collected by sensors with different sampling periods to the same time point for time synchronization;

[0092] The Lagrangian interpolation method includes: using a constant acceleration model to predict the radar data, then using the camera time as the input of the Lagrangian interpolation method, and using the Lagrangian interpolation formula to interpolate the target parameters measured by the radar at time k-1, time k, and time k+1, so as to obtain the same radar measurement value as the camera time.

[0093] The radar sampling period is faster than the camera sampling period, so the camera sampling period is used as the benchmark.

[0094] 3. By using an association matching algorithm, it is determined whether the data measured by different sensors belong to the same target. This embodiment uses the GNN algorithm;

[0095] The GNN algorithm includes: 1) using the camera measurement values ​​and radar measurement values ​​after spatiotemporal synchronization as the input of the GNN algorithm; 2) using the designed weighting function formula To calculate the performance index, where e represents the error; 3) When the performance index meets the threshold, it is determined that the data measured by different sensors are the same target;

[0096] 4. Perform fusion tracking. Target fusion is when the state parameters of the same target are provided by different sensors at the same time, and there is a certain deviation between the two sets of parameters. Therefore, a fusion algorithm is needed to fuse the target parameters measured by different sensors. This embodiment uses the Kalman filter algorithm to fuse the target; tracking is to estimate the future motion state of the target by using the target's motion model and the target's observation data. The target motion model used in this embodiment is a constant acceleration motion model, and the state quantity is (X, Y, V x 、V y , A X , A y ) T , where X is the longitudinal distance, Y is the lateral distance, Vx is the longitudinal velocity, Vy is the lateral velocity, Ax is the longitudinal acceleration, and Ay is the lateral acceleration;

[0097] Kalman filter algorithm processing: By using the time-synchronized camera measurements (X, Y, V x 、V y , A X , A y ) T , radar measurement values ​​(X, Y, V x 、V y, A X , A y ) T As the input of the Kalman filter, the target state is predicted; then the error covariance matrix P is updated using the state matrix F; the Kalman filter gain K is calculated using the error covariance matrix P, the observation matrix H, and the covariance matrix R of the sensor noise; (X, Y, V x 、V y , A X , A y ) T Update the measurement value X; finally update the covariance error P.

[0098] 5. Output of dangerous targets. By using the rule-based algorithm and the fusion results of different sensor targets currently obtained, the most dangerous target type and target state are screened out from these targets, and the target state is converted into an absolute coordinate system, providing important reference information for subsequent early warning units and perception units.

[0099] For the roadside perception algorithm, this embodiment only considers the target-level fusion perception algorithm of the camera and millimeter-wave radar similar to the vehicle side, so the roadside camera and millimeter-wave radar and other sensors are required to provide target-level data of the detection results. Existing roadside cameras generally do not have the function of outputting target-level results. The target-level results refer to information such as the target position and speed, so a corresponding image detection algorithm is required to achieve the output of target-level results. This embodiment meets the camera image processing requirements and requires reasonable configuration of hardware and algorithm software to ensure that the image processing delay of the roadside perception algorithm is controlled within 70ms.

[0100] In view of the characteristics of roadside cameras and the delay in single-target and multi-target detection, the maximum delay value is selected according to the single-target delay time and the multi-target delay time, and the external target detection data is unified to ensure the consistency of data reading.

[0101] Under the above premise, the data received by roadside fusion perception is the target-level result data output by the roadside camera and millimeter-wave radar, including target ID and type, location, speed and other information.

[0102] The roadside fusion process includes:

[0103] For the conventional fusion perception algorithm process, data preprocessing is required first to eliminate empty targets and invalid targets; referring to the vehicle-side fusion method, absolute time interpolation matching is used to synchronize the camera perception results with the radar perception results; similar to time alignment, spatial alignment also uses absolute position information. This part of the timestamp information and longitude and latitude positioning information can be provided by the roadside unit, or the longitude and latitude coordinates of the roadside unit can be set according to the absolute positioning accuracy when the roadside system is installed and calibrated, but the system as a whole needs to meet the accuracy requirements (RTK positioning is used as the position reference). The timestamp information is triggered by the roadside millimeter wave to trigger the camera shutter to ensure that the time of the two devices is consistent, and the timestamp uses the sensor time. Spatial synchronization is completed through the roadside absolute positioning information. After completing the spatiotemporal synchronization, GNN association matching and Kalman filter weighted fusion are still used to finally output the fused target information. The target after the roadside fusion processing is the same as the target after the vehicle-side fusion, mainly including timestamp, dangerous target ID, coordinates, relative speed, relative acceleration, etc.

[0104] The target result after roadside fusion processing needs to be packaged into corresponding formatted data in the edge computing node according to the format requirements of the roadside unit, and sent to the roadside unit using the communication protocol specified by the roadside unit, and broadcast by the roadside unit. The on-board unit receives the broadcast information of the roadside unit within the communication range, and transparently transmits the relevant information to the vehicle controller for subsequent vehicle-road fusion perception processing.

[0105] For the analysis of the on-board controller on the on-board unit, data analysis is performed according to the analysis method provided by the supplier to ensure that the on-board controller can parse, read and process the data of the on-board unit through the developed analysis program.

[0106] The input of the vehicle-road perception fusion algorithm is the final vehicle-side fusion perception result after vehicle-side fusion perception processing and the final road-side fusion perception result after road-side fusion perception processing. In the vehicle-side and road-side fusion algorithms, the final output results have been processed by various fusion perception algorithm modules into absolute longitude and latitude coordinate data and speed in the world coordinate system, that is, spatial alignment has been completed. The vehicle-road fusion perception algorithm needs to complete time alignment processing and subsequent association matching, Kalman filter fusion, and multi-target matching tracking.

[0107] For the vehicle-side and roadside perception results, due to the different transmission topologies, and taking into account the many links involved in roadside camera image processing, perception fusion, RSU-OBU data communication and OBU transparent data analysis, a large time delay will inevitably occur. The preliminary estimate of this embodiment is that it may reach 0.2s.

[0108] The vehicle-side fusion perception information and the roadside fusion perception information are based on the earth coordinates, and no spatial synchronization is required. In terms of time, due to the time consumption of multiple processes such as algorithms and signal transmission, the roadside fusion perception information will have a certain delay relative to the vehicle-side fusion perception information. For the roadside fusion perception information that always has a certain lag, this embodiment tests and statistics the time delay, and makes a preliminary extrapolated prediction of the roadside perception target list according to the specific time delay size, compensates for the impact of the delay, and then performs subsequent vehicle-road fusion algorithm processing.

[0109] The vehicle-road fusion algorithm is consistent with the vehicle-side and road-side fusion perception. It uses GNN association matching, Kalman weighted fusion and GNN multi-target matching in three steps for fusion processing. The specific algorithm is the same as the description of the vehicle-side perception algorithm. The only difference is that the two parts of the fused information are changed from the information of the vehicle-side camera and millimeter-wave radar to the information after vehicle-side fusion and the information after road-side fusion.

[0110] After vehicle-road fusion processing, the information of the target matched by the vehicle and the road can be obtained. At the same time, the unmatched vehicle-side targets and roadside targets are retained. These three parts are superimposed as the final vehicle-road fusion perception information list and sent to the early warning unit for target behavior prediction and decision processing.

[0111] Example 2

[0112] Combined with Example 1 and Figure 4 As shown, this embodiment further provides a vehicle early warning control system, which implements the steps of the vehicle early warning control method described in any one of the embodiments 1, including:

[0113] An acquisition unit, used for acquiring TSR identification information;

[0114] The first judgment unit is used to judge whether the TSR identification information belongs to an activation scenario for activating the vehicle warning function; this embodiment provides three activation scenarios, namely:

[0115] Scenario 1: Traffic lights are blocked during TSR recognition;

[0116] Scenario 2: Traffic lights are blurred or easily confused during TSR recognition;

[0117] Scenario 3: Traffic signs are blurred or easily confused during TSR recognition.

[0118] When it is judged to belong to any of these three scenarios, the scene flag is activated for the vehicle and it enters the perception area to obtain vehicle-road perception fusion information.

[0119] A perception unit, used to obtain vehicle-road perception fusion information;

[0120] The second judgment unit is used to judge whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information;

[0121] The early warning unit is used to control the vehicle to issue early warning instructions to the driver.

[0122] Furthermore, the sensing unit includes:

[0123] Roadside unit, used to obtain roadside end fusion perception information;

[0124] The vehicle-mounted unit is used to obtain vehicle-side fusion perception information.

[0125] Furthermore, the second determining unit includes:

[0126] When the vehicle activates scenario 1 or scenario 2, the roadside unit broadcasts traffic light status information, which may be green or red / yellow. At this time, the second judgment unit determines whether the vehicle can pass the intersection safety line normally or whether it is necessary to issue a warning instruction to the driver based on the vehicle-road perception fusion information;

[0127] When the vehicle activation scenario is scenario 3, the roadside unit broadcasts traffic sign information or other specific areas where the speed reduction is required. At this time, the second judgment unit determines whether the vehicle is driving normally or whether a warning instruction needs to be issued to the driver based on the vehicle perception fusion information;

[0128] If a warning instruction is issued to the driver, the vehicle will drive normally when it meets the safety line of the intersection where the vehicle can drive normally or the speed meets the speed limit requirements.

[0129] This embodiment monitors traffic light status information or traffic sign information in real time through a roadside unit and publishes traffic light information and traffic sign information to determine whether the vehicle's current driving status can safely pass the traffic light or traffic sign position, and then issues warning instructions at appropriate locations to achieve the purpose of safe driving.

[0130] The vehicle warning triggering principle in this embodiment is: when the vehicle enters the communication range of the roadside unit and is still a certain distance away from the traffic lights and traffic signs, it can be guaranteed that after the warning instruction is issued, the driver has time and can make the vehicle safely pass the traffic lights or traffic signs through appropriate operating behaviors.

[0131] The warning distance is calculated by using formula (1), formula (2) and formula (3) in Example 1.

[0132] The trigger conditions for the warning include:

[0133] The vehicle enters the communication range of the roadside unit;

[0134] The current vehicle speed cannot safely pass the traffic light / traffic sign;

[0135] The distance between the vehicle and the traffic light / traffic sign is less than the warning distance calculated above.

[0136] The control involved in this embodiment is mainly the control of HMI, that is, the control of the warning function. When the vehicle needs to perform driving warning, the warning unit sends a corresponding control signal to the HMI interface, and gives the driver a warning instruction with sound and image.

[0137] When the driver takes over and exits the warning mode, for example, in a gentle braking stop state, when the driver performs a braking operation but the deceleration is insufficient, the warning unit continues to issue warning instructions to the driver and continues to calculate the time to pass the road conditions ahead.

[0138] Example 3

[0139] Combined with Example 2 and Figure 5 As shown, this embodiment provides a vehicle early warning implementation method, including:

[0140] The vehicle has three pre-stored activation scenarios, namely, scenario 1, scenario 2 and scenario 3;

[0141] The warning system uses TSR identification information to determine whether the vehicle has activated any of the above three scenario flags. If so, it proceeds to the next step, otherwise the process ends;

[0142] After the warning system activates the relevant scene flag, the vehicle enters the perception area, performs vehicle-road perception fusion, and obtains vehicle-road perception fusion information;

[0143] According to the activated scene flag and the vehicle-road perception fusion information, when the roadside unit releases the traffic light status information, it determines whether the traffic light is green or red / yellow, and then determines whether the vehicle can safely pass the intersection safety line when the light is green. If yes, the vehicle drives normally, otherwise it issues a warning instruction to the driver; when the roadside unit releases the traffic sign information, it determines whether the current vehicle speed meets the sign requirements according to the traffic sign for speed limit or other specific area signs that require speed reduction. If yes, the vehicle drives normally, otherwise it issues a warning instruction to the driver; after the warning instruction is issued, the driver makes corresponding operations according to the instruction and after event A or event B, the vehicle resumes normal driving. Event A means that the vehicle can pass the intersection safety line when the traffic light is green, and event B means that the vehicle speed does not exceed the maximum speed limit of the traffic sign.

[0144] Example 4

[0145] In combination with Example 1 and Example 2, this embodiment provides a vehicle that implements the steps of the vehicle warning control method described in any one of Example 1.

[0146] The technical solution provided by the present invention can enable the vehicle to obtain the traffic information status ahead in advance, perceive and judge the traffic environment in advance, and determine whether it is necessary to issue warning instructions to the driver based on the judgment, such as alarm, active deceleration or emergency braking and other related warning instructions, so that the vehicle can achieve the purpose of safe driving.

[0147] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle early warning control method, Features: The following steps are involved: S1: Obtain TSR identification information and determine whether the TSR identification information belongs to the activation scenario for activating the vehicle warning function. If so, execute S2, otherwise terminate the process; S2: After the vehicle warning function is activated, the vehicle enters the perception area to perform vehicle-road perception fusion to obtain vehicle-road perception fusion information; S3: Determine whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information. If necessary, control the vehicle to issue a warning instruction to the driver. If not, terminate the process. The activation scenarios include: Traffic lights are blocked during TSR recognition; Traffic lights are blurred or easily confused when TSR is used for recognition; When TSR recognizes traffic signs that are blurry or easily confused; S3 includes: obtaining a judgment result when judging whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information; The judgment results include: When the traffic light ahead is green, calculate whether the vehicle can safely pass the intersection safety line based on the current vehicle motion state; When the traffic light ahead is red or yellow, the vehicle warning distance is calculated based on the current vehicle motion status; When the traffic sign ahead indicates a speed limit or other specific area where the speed reduction is required, the current vehicle motion state is combined to calculate whether the vehicle exceeds the maximum speed limit; After determining the direction of the vehicle's current lane, the traffic light status is determined through V2X information. When the traffic light is red or yellow, the vehicle's motion status and traffic light information are comprehensively determined to obtain the vehicle's current warning distance. When the vehicle enters the warning distance, the driver is reminded. After receiving the reminder, the driver takes the initiative to brake and decelerate. The warning distance of the vehicle is calculated by the following formula: Among them, v sv is the initial vehicle speed, t svd ,t rbr and the driver reaction time and the braking system response time, respectively. 1 is the mild braking deceleration, which is a vector value, acceleration is a positive value and deceleration is a negative value; d pre To warn of safe distance; The S2 includes: The vehicle-road perception fusion information includes a combination of vehicle-side perception fusion information and roadside perception fusion information; The target after vehicle-side fusion processing mainly contains information including timestamp, target ID, coordinates, relative speed V x 、V y , acceleration A x ; When the vehicle controller receives the target perception result on the vehicle side, it first calibrates the spatial position of the camera and radar, and then uses the time synchronization method based on Lagrange interpolation to synchronize the camera and radar data; After the camera and radar data are synchronized in time and space, the global nearest neighbor (GNN) algorithm is used to match the target. When the performance function E of the GNN algorithm is within the threshold range of 0 to Eth and the E value is the smallest, the two are considered to be the same target. Then the Kalman filter method is used to track the data of the matched targets and the targets that are not matched by the camera / radar, and the weighted method is used to fuse the matched target data; Finally, the CIPV / P targets, the targets that are not matched by the camera, the CIPV / P targets converted to absolute positions, and the parameters of the targets that are not matched by the camera converted to absolute positions are selected as the input of the perception unit and the early warning unit.

2. A vehicle early warning control method according to claim 1, Features: When the traffic light ahead is green, calculating whether the vehicle can safely pass the intersection safety line in combination with the current vehicle motion state includes: If the vehicle cannot safely pass the intersection safety line, the control vehicle will issue a warning instruction to the driver; If the vehicle can safely pass the intersection safety line, the process is terminated and the driver drives normally.

3. A vehicle early warning control method according to claim 1, Features: When the traffic sign ahead indicates a speed limit or other specific area where the speed reduction is required, the calculation of whether the vehicle exceeds the maximum speed limit is performed in combination with the current vehicle motion state, including: If the vehicle does not exceed the maximum speed limit, the control warning system will issue normal driving instructions to the driver; If the vehicle exceeds the maximum speed limit, the warning system will be controlled to issue a driving warning instruction to the driver.

4. A vehicle early warning control system, implementing the steps of the vehicle early warning control method according to any one of claims 1 to 3, Features: include: An acquisition unit, used for acquiring TSR identification information; A first judgment unit, used to judge whether the TSR identification information belongs to an activation scenario for activating a vehicle warning function; A perception unit, used to obtain vehicle-road perception fusion information; The second judgment unit is used to judge whether it is necessary to issue a warning instruction to the driver through the vehicle-road perception fusion information; The early warning unit is used to control the vehicle to issue early warning instructions to the driver.

5. A vehicle early warning control system according to claim 4, Features: The sensing units include: Roadside unit, used to obtain roadside end fusion perception information; The vehicle-mounted unit is used to obtain vehicle-side fusion perception information.

6. A vehicle, Features: Implement the steps of the vehicle early warning control method described in any one of claims 1-3.