Intelligent car following method and system

Abstract

The invention provides an intelligent vehicle following method and system, and the method comprises the following steps: obtaining the driving environment information of a vehicle in real time, carrying out the visual display, and recognizing a driving region and a vehicle which can be followed in the displayed driving environment information of the vehicle; receiving a trigger instruction, and taking the selected vehicle capable of being followed as a target following vehicle; automatically generating a vehicle-following strategy according to the position information of the lane where the target vehicle-following vehicle is located, the vehicle on the lane, the traffic marker and the obstacle information; and a corresponding control instruction is generated based on the vehicle following strategy to automatically control the vehicle to execute acceleration and deceleration, lane changing and overtaking processing, and vehicle following driving of the vehicle relative to the target vehicle following vehicle is completed. According to the method, the travelable area and the vehicle which can be followed are divided for the user to select, in the process, the user only needs to give the vehicle selection instruction which can be followed, the vehicle which needs to be followed is selected, the system follows the generated moving track to automatically complete vehicle following, and the vehicle following mode is more flexible.

Description

Technical Field

[0001] The present invention belongs to the technical field of assisted driving, and in particular relates to an intelligent vehicle following method and system. Background Art

[0002] With the rapid development of science and technology, intelligent driving has gradually moved from concept to reality and has become an important development direction of the modern automobile industry. Nowadays, more and more cars are equipped with advanced L2+ driving assistance functions, among which high-speed navigation assisted driving (NOA) has become the core configuration of many smart cars with its outstanding performance on highways, bringing unprecedented convenience and comfort to drivers.

[0003] However, in actual use, the active lane changing function of high-speed NOA has caused a lot of controversy. Through data analysis of major automobile evaluation platforms, no matter what brand and model of the vehicle, it is inevitable that users will question the timing of active lane changes. The root cause of these problems is that each driver has his or her unique driving habits and preferences, and the high-speed NOA system currently on the market only provides three preset modes of "smooth", "standard" and "aggressive" in terms of active lane change control. These modes mainly achieve different lane changing strategies by adjusting the compression ratio and the compression duration. Only when the vehicle's driving state meets specific parameter conditions will the active lane change operation be triggered. Although this control method based on fixed parameter rules simplifies the system design to a certain extent, it cannot fully adapt to complex and changeable driving scenarios and personalized user needs. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent vehicle following method and system, which solves the problem that the vehicle following method in the related art is not flexible enough.

[0005] To this end, in a first aspect, the present invention provides an intelligent vehicle following method, comprising the following steps:

[0006] Acquire the vehicle's driving environment information in real time and display it visually, wherein the driving environment information includes lanes, vehicles on the lanes, traffic signs, and obstacle information;

[0007] Identifying a drivable area and a following vehicle in the displayed driving environment information of the vehicle;

[0008] receiving a trigger instruction, and triggering the selected following vehicle as a target following vehicle;

[0009] Automatically generate a following strategy based on the lane where the target following vehicle is located, the location information of vehicles on the lane, traffic signs and obstacle information;

[0010] Based on the following vehicle strategy, corresponding control instructions are generated to automatically control the vehicle to execute acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle.

[0011] Optionally, dividing the drivable area and the following vehicles according to the environmental information includes:

[0012] According to the location information of lanes, vehicles on lanes, traffic signs, and obstacles, the movable path of the vehicle is calculated through a path planning algorithm; the drivable area and the vehicles that can be followed are divided according to the movable path of the vehicle.

[0013] Optionally, a moving path for the self-vehicle to move behind the target vehicle within the drivable area is obtained based on the position of the target following vehicle relative to the self-vehicle, and the self-vehicle moves behind the target vehicle along the moving path and follows the target vehicle.

[0014] Optionally, when the target following vehicle is the first vehicle in front of the own vehicle in the lane, the own vehicle speed is adjusted to move behind the target vehicle to follow the vehicle.

[0015] Optionally, when the target following vehicle is the vehicle in front of an adjacent lane, it is determined whether there is safe lane change space at present and within a safe time based on the position and speed of the vehicles in the adjacent lane; if there is safe lane change space, the self-vehicle changes lanes and moves to the rear of the target lane to follow the vehicle.

[0016] Optionally, when the target following vehicle is the Nth vehicle in front of the own vehicle in the lane, it is determined whether the overtaking condition is met. If the overtaking condition is met, the own vehicle changes lanes to an adjacent lane, accelerates and changes lanes again to the rear of the target vehicle, where N is greater than 1.

[0017] In a second aspect, an intelligent vehicle following system is provided, which is used for the intelligent vehicle following method, and includes:

[0018] A data acquisition module, the data acquisition module is used to collect driving environment information; wherein the driving environment information includes lanes, vehicles on the lanes, traffic signs, and obstacle information;

[0019] The cockpit host is used to interact with the outside world, perform visual display and receive trigger instructions;

[0020] A controller is communicatively connected with the data acquisition module and the cockpit host, and is used for visually displaying the driving environment information after fusing it through the cockpit host, and generating a following strategy according to a trigger instruction fed back by the cockpit host, and generating corresponding control instructions based on the following strategy to automatically control the vehicle to execute acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle.

[0021] Optionally, the controller includes a perception fusion module, a fusion positioning module, a map data engine module and an environmental model module. The fusion positioning module is used to fuse the driving environment information collected by the data acquisition module to output the coordinates of the vehicle and the relative position in the lane; the map data engine module is used to match the map-related information according to the coordinates of the vehicle in combination with the navigation path, and output map information, real-time navigation path information and positioning information; the environmental model module is used to extract the map semantic level, road topology structure and navigation information, and construct a static global path; the perception fusion module is used to perform target recognition on the environmental information collected by the data acquisition module.

[0022] Optionally, the controller further includes a decision-making planning module, which is used to dynamically generate a movement trajectory in real time according to information from the perception fusion module and the environment model module.

[0023] Optionally, the controller further includes a control module, and the control module is used to output control instructions according to the movement trajectory.

[0024] Beneficial effects:

[0025] (1) The present disclosure provides an intelligent vehicle following method and system, which divides a drivable area and vehicles that can be followed for selection by a user. After the user gives a vehicle selection instruction, the selected vehicle that can be followed is used as a target vehicle according to the user's vehicle selection instruction, and moves to the rear of the target vehicle according to a planned moving route and follows the vehicle. During the process, the user only needs to give a vehicle selection instruction and select a vehicle that can be followed. The system automatically follows the generated moving trajectory to complete the following, and the vehicle following method is more flexible.

[0026] (2) In the present disclosure, whether the following condition is met is determined based on the position of the target vehicle relative to the own vehicle, and a moving path is automatically generated for the own vehicle to move to the rear of the target vehicle within the drivable area, thereby performing accelerated following, lane changing following, or overtaking following, thereby shortening the speed from triggering the following to completing the following while maintaining safety.

[0027] (3) The perception fusion module, fusion positioning module, map data engine module and environment model module in the present disclosure cooperate to construct a restored world, ensure the accuracy of information such as lane position, type, vehicles in the lane, obstacles, etc., provide a basis for the decision-making planning module to generate a moving trajectory through route planning, and ensure the accuracy of the moving trajectory.

[0028] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 A method flow chart of an embodiment of an intelligent vehicle following method in the present disclosure;

[0031] Figure 2 This is a diagram showing a car following interface of an embodiment of an intelligent car following method in the present disclosure;

[0032] Figure 3 A system structure diagram of another embodiment of an intelligent vehicle following method in the present disclosure;

[0033] In the figure, 100-data acquisition module, 101-extracabin sensor, 102-combined positioning P-Box, 200-cockpit host, 201-world restoration module, 202-following target status display module, 300-controller, 301-perception fusion module, 302-fusion positioning module, 303-map data engine module, 304-environmental model module, 305-decision planning module, 306-control module, 400-actuator. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] The terms "first", "second", "third", "fourth", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances. For example, without departing from the scope of this document, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0036] The word "if" as used herein may be interpreted as "when" or "when" or "in response to determining," depending on the context.

[0037] Furthermore, as used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context indicates otherwise.

[0038] It should be further understood that the terms “comprises” and “includes” indicate the existence of features, steps, operations, elements, components, items, types, and / or groups, but do not exclude the existence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, types, and / or groups.

[0039] The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". An exception to this definition will only occur when a combination of elements, functions, steps or operations are inherently mutually exclusive in some manner.

[0040] With the rapid development of science and technology, intelligent driving has gradually moved from concept to reality and has become an important development direction of the modern automobile industry. Nowadays, more and more cars are equipped with advanced L2+ driving assistance functions, among which high-speed navigation assisted driving (NOA) has become the core configuration of many smart cars with its outstanding performance on highways, bringing unprecedented convenience and comfort to drivers.

[0041] However, in actual use, the active lane changing function of high-speed NOA has caused a lot of controversy. Through data analysis of major automobile evaluation platforms, no matter what brand and model of the vehicle, it is inevitable that users will question the timing of active lane changes. The root cause of these problems is that each driver has his or her unique driving habits and preferences, and the high-speed NOA system currently on the market only provides three preset modes of "smooth", "standard" and "aggressive" in terms of active lane change control. These modes mainly achieve different lane changing strategies by adjusting the compression ratio and the compression duration. Only when the vehicle's driving state meets specific parameter conditions will the active lane change operation be triggered. Although this control method based on fixed parameter rules simplifies the system design to a certain extent, it cannot fully adapt to complex and changeable driving scenarios and personalized user needs.

[0042] To this end, in a first aspect, the present disclosure provides Figure 1 An intelligent vehicle following method is shown, comprising the following steps: performing the following steps under the condition of activating high-speed NOA (Navigate on Autopilot, navigation-assisted driving), wherein high-speed NOA is used to provide accurate route planning and real-time road condition analysis to assist the driver in driving and improve driving efficiency when the vehicle is traveling on the highway.

[0043] S100, real-time acquisition of vehicle driving environment information and visual display,

[0044] Among them, the driving environment information includes lanes, vehicles on lanes, traffic signs, and obstacle information; environmental data is collected through sensors such as cameras and radars in the data acquisition module, and the environmental data is restored to the real world environment through the controller and presented to the driver in a visual manner, helping the driver to better understand the vehicle's surrounding conditions and improve driving safety and convenience.

[0045] In this embodiment, a three-lane four-line model is used to display the lane lines and road conditions of the self-lane and adjacent lanes. For example, the vehicle information 100m ahead, 60m behind, and 20m to the side is displayed in real time. At the same time, static obstacles such as water barriers, cone barrels, pile barrels and other targets are perceived through visual sensors such as cameras. The distance of the target object is displayed in real time when the distance is less than 80cm by sensing and measuring distance through distance sensors such as radar. By restoring the world display, it can help drivers better understand the vehicle's driving environment and the decision-making basis of the intelligent driving system, thereby improving driving safety and comfort.

[0046] S200, identifying a drivable area and a following vehicle in the displayed driving environment information of the vehicle;

[0047] This includes calculating the vehicle's movable path through a path planning algorithm based on the location information of lanes, vehicles and obstacles; and dividing the drivable area and followable vehicles based on the vehicle's movable path.

[0048] The relative position relationship between the vehicle or obstacle and the ego vehicle is calculated based on the position information of the lane, vehicle and obstacle, and the movable path of the ego vehicle is calculated within the lane using the path planning algorithm. The space of all movable paths constitutes the drivable area, and the vehicles that can reach and follow the ego vehicle through the movable path are regarded as the following vehicles.

[0049] In one embodiment, a path planning algorithm is used to divide the drivable space based on the relative position relationship between the vehicle or obstacle and the vehicle every 50ms, wherein the path planning algorithm is a combination of the A* algorithm and the Freespace algorithm. The A* algorithm is a path search algorithm that maintains an open list and a closed list during the search process, and continuously selects the node with the smallest value for expansion until the target node is found or all possible nodes are traversed. Since the high-speed navigation function does not support long-term cross-lane riding, the A* algorithm can exclude the target position that can only be reached by long-term riding, thereby ensuring the accuracy of the path search. The Freespace algorithm is mainly used to deal with the obstacle problem in path planning. It divides the space in the map into free space and obstacle space, and finds a feasible path by searching the free space. Usually, the map is preprocessed first to mark which areas are passable free space and which are inaccessible obstacle areas. In the process of combining the A* algorithm with the Freespace algorithm, in the search process of the A* algorithm, each time a node is expanded, it is necessary to determine whether the node is reachable according to the space divided by the Freespace algorithm. Only nodes in the free space are expandable nodes, and nodes in the obstacle space are directly excluded. For example, when the A* algorithm tries to expand a node, it first checks whether the node is in free space through the preprocessing information of the Freespace algorithm. If it is, it is added to the open list for evaluation and further expansion; if not, it ignores the node and continues to check other nodes. In this way, the drivable area and the following vehicles are divided in turn, which improves the search efficiency, enhances the path safety, and adapts to complex map environments.

[0050] like Figure 2 As shown, in the visual image displayed in the restored world, the blue wireframe is the position of the vehicle, the vertical dotted line is the lane line, the gray wireframe is the vehicle that can be selected in front, and the green wireframe is the drivable area. The range of the current vehicle that can be followed and the drivable area are displayed in real time.

[0051] S300, receiving a trigger instruction, and triggering the selected vehicle that can be followed as a target vehicle to follow;

[0052] Among them, the following vehicle selection instruction is issued by the driver. In one embodiment, the driver can select a following vehicle as the target vehicle in the display screen in the restored world display, which is deemed to have issued an eradicable vehicle selection instruction, and the voice confirmation "The target has been selected as the following target, and the intelligent following function is activated" is turned on. At the same time, the corresponding indicator light on the instrument panel lights up, the interface highlights the target vehicle, and displays its core parameters such as speed and distance, and displays the target location box and planned path. When the driver clicks on the target vehicle again, the target vehicle can be cancelled and the following process can be stopped.

[0053] S400, automatically generating a following vehicle strategy according to the lane where the target following vehicle is located, the position information of the vehicles on the lane, the traffic signs and the obstacle information;

[0054] Among them, the moving path of the self-vehicle to move to the rear of the target vehicle in the drivable area is obtained according to the position of the target vehicle relative to the self-vehicle, and when the moving conditions are met, the self-vehicle moves to the rear of the target vehicle through the moving path and follows the target vehicle.

[0055] According to the position of the target vehicle relative to the self-vehicle, it can be divided into three situations: following the vehicle in front, following the vehicle after changing lanes, and following the vehicle after overtaking.

[0056] S401. When the target vehicle is the first vehicle in front of the own vehicle in the lane, adjust the own vehicle speed to move behind the target vehicle to follow the vehicle.

[0057] Among them, when the target vehicle is the first vehicle in front of the own vehicle in the lane, the vehicle speed is adjusted in real time according to the time interval set by the current high-speed NOA and the relative speed between the target vehicle and the own vehicle. If the actual distance is less than the safe distance, the vehicle performs braking control according to the relative speed to increase the distance with the target vehicle. If the actual distance is less than the safe distance, the vehicle performs acceleration control according to the relative speed to reduce the distance with the target vehicle. The intensity of braking or acceleration is determined according to the distance difference and the relative speed.

[0058] S402: When the target vehicle is the vehicle in front of the adjacent lane, determine whether there is safe lane change space at present and within the safe time based on the position and speed of the vehicles in the adjacent lane; if there is safe lane change space, the self-vehicle changes lanes and moves to the rear of the target lane to follow the vehicle.

[0059] Among them, when the target vehicle is the vehicle in front of the adjacent lane, the safe lane change space required for the vehicle to change lanes is calculated based on the position of the vehicle in the adjacent lane. The safe lane change space is at least greater than the length of the vehicle plus the safety margin, and the safety margin is 1-2m. At the same time, according to the position and speed of the vehicle in the adjacent lane, it is calculated whether there is still safe lane change space within the safety time, where the safety time is the time within the next 3-5s.

[0060] At the same time, the visual sensor perceives the road divergence and merging points as well as the virtual and real change points to determine whether there is a collision or regulatory risk in the current lane change scenario.

[0061] When all lane change conditions are met, a lane change path is planned based on the vehicle dynamics characteristics so that the vehicle can move from the current lane to the lane where the target vehicle is located. In one embodiment, the lane change path uses a quadratic polynomial y=ax 2 +bx+c is used for real-time planning. By adjusting the coefficients a, b, and c, lane change trajectories of different shapes are obtained to ensure that the vehicle can smoothly transition from the current lane to the target lane during the lane change process.

[0062] When the vehicle successfully changes lanes to the target lane, the speed is intelligently adjusted to follow the target vehicle based on its relative speed and time distance.

[0063] If the lane change condition is not met, and the duration of the lane change condition not being met is greater than a set threshold, the lane change is canceled and a lane change cancellation prompt is fed back. In one embodiment, the set threshold is 10s.

[0064] If the target vehicle changes lanes to another lane while the ego vehicle is initiating a lane change, the ego vehicle changes lanes to the lane where the target vehicle was initially located and maintains cruising.

[0065] S403. When the target vehicle is the Nth vehicle in front of the own vehicle in the lane, determine whether the overtaking condition is met. If the overtaking condition is met, the own vehicle changes lanes to an adjacent lane, accelerates and changes lanes again to the rear of the target vehicle, where N is greater than 1.

[0066] Among them, the overtaking conditions include:

[0067] Scene judgment conditions: Based on the environmental information, the system determines whether the vehicle is in a safe overtaking scenario. For example, tunnels, sharp bends, 1.5 km before the off-ramp, and 2 km before and after the toll station are all unsafe overtaking scenarios and overtaking is not allowed. If the scene judgment conditions are not met, the vehicle will enter the overtaking and following cancellation state.

[0068] Space judgment condition: judge the position and speed of the vehicle in the current lane and the adjacent lane based on the environmental information, and judge the lane change and overtaking space in the lane on the left of the vehicle. The lane change and overtaking space is greater than the longitudinal distance from the vehicle to the first vehicle in front of the vehicle plus the vehicle body length plus the safety distance, where the safety distance is 3-5m. If the lane change and overtaking space cannot be met, the vehicles in the adjacent lanes are too dense to overtake, and the space judgment condition is not met, and the overtaking and following vehicle cancellation state is entered.

[0069] Speed ​​judgment condition: judge the relative speed between the vehicle and the first vehicle in front of the vehicle based on the environmental information, and judge whether the vehicle can overtake and change lanes within the safe overtaking time and at the safe overtaking acceleration within the lane change space. In one embodiment, the safe overtaking time is set to 10s and the safe overtaking acceleration is set to 4m / s 2 If the overtaking cannot be completed within the safe overtaking time, or the acceleration required during acceleration exceeds the safe overtaking acceleration, or the speed during overtaking exceeds the speed limit, the speed judgment condition is not met and the overtaking and following vehicle cancellation state is entered.

[0070] When all overtaking conditions are met, overtaking and following is triggered. The vehicle first changes lanes to the adjacent lane on the left, then accelerates to overtake, then changes lanes to the right to return to the original lane, and follows the target vehicle.

[0071] When the vehicle changes lanes to the adjacent lane on the left, the lane change conditions must be met, including:

[0072] a. There is safe lane-changing space in the adjacent lane. The safe lane-changing space is at least greater than the length of the vehicle plus a safety margin, which is 1-2 meters. Collisions can be avoided by reserving safe lane-changing space.

[0073] b. The lanes in the safe lane change space are dotted lines. You can change lanes when the lanes in the safe lane change space are dotted lines.

[0074] c. No collision risk within the safety time. Calculate whether there is still safe lane change space within the safety time based on the position and speed of the vehicles in the adjacent lanes. The safety time is the time within the next 3-5 seconds. That is, it is predicted that there will be no trajectory intersection between the vehicle and the vehicles in the adjacent lanes within the safety time to avoid collision.

[0075] If the lane change conditions are met, the vehicle will change lanes to the adjacent lane on the left, and detect in real time whether there is a collision risk in the surrounding environment. If there is a collision risk, the vehicle will slow down and avoid it in time and issue a prompt to the driver. If there is no collision risk, the vehicle will change lanes quickly.

[0076] After entering the left lane, the relative distance and speed between the vehicle and the target vehicle and the first vehicle in front of the vehicle are calculated in real time, and a stable acceleration rate is maintained at a safe acceleration rate. The speed of vehicles in adjacent lanes is continuously monitored to avoid rear-end collisions.

[0077] When the original lane meets the lane change conditions, the vehicle triggers a lane change to the right, moves back to the original lane, and drives behind the target vehicle.

[0078] When the vehicle changes lanes to the right lane, the lane change conditions must be met, including:

[0079] a. There is safe lane-changing space in the original lane. The safe lane-changing space is at least greater than the length of the vehicle plus the safety margin, which is 1-2 meters. Collisions can be avoided by reserving safe lane-changing space.

[0080] b. The lanes in the safe lane change space are dotted lines. You can change lanes when the lanes in the safe lane change space are dotted lines.

[0081] c. No collision risk within the safety time. Calculate whether there is still safe lane change space within the safety time based on the position and speed of the vehicle in the original lane. The safety time is the time within the next 3-5 seconds. That is, it is predicted that there will be no trajectory intersection between the vehicle and the vehicle in the original lane within the safety time to avoid collision.

[0082] If the lane change conditions are met, the vehicle changes lanes to the right of the original lane and detects in real time whether there is a collision risk in the surrounding environment. If there is a collision risk, it will slow down and avoid it in time and issue a prompt to the driver. If there is no collision risk, it will change lanes quickly. When the vehicle enters the center of the original lane and completes the lane change, it will slow down smoothly and follow the target vehicle.

[0083] During the process of overtaking and changing lanes, if the lane changing condition is not met, and the duration of the lane changing condition not being met is greater than a set threshold, the lane changing is canceled and a lane changing cancellation prompt is fed back. In one embodiment, the set threshold is 5 seconds.

[0084] When the ego vehicle is overtaking and changing lanes, if the target vehicle leaves or changes lanes to another lane, the ego vehicle changes lanes to the lane where the target vehicle was originally located and adjusts the speed to the ego vehicle's initial speed.

[0085] As shown in the following table, when the ego vehicle passes through the drivable area and moves to the rear of the target vehicle and follows the vehicle, the following status is fed back to the driver in real time.

[0086]

[0087]

[0088] S500. Generate corresponding control instructions based on the following vehicle strategy to automatically control the vehicle to execute acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle.

[0089] According to the control command, the ego vehicle moves to the rear of the target following vehicle according to the moving trajectory and follows the vehicle.

[0090] In a second aspect, an intelligent vehicle following system is provided, which is used to implement an intelligent vehicle following method, including:

[0091] The data acquisition module 100 is used to collect driving environment information; wherein the driving environment information includes lanes, vehicles on the lanes, traffic signs, and obstacle information;

[0092] The data acquisition module 100 includes an extravehicular sensor 101 and a combined positioning P-Box 102. The extravehicular sensor 101 is composed of a variety of different sensors. In one embodiment, the extravehicular sensor 101 is configured as 6V5R12U, that is, 1 monocular front-view camera, 1 monocular rear-view camera, 4 side-view cameras, 1 front radar, 4 corner radars and 12 ultrasonic radars, which can sense environmental information and dynamic and static targets in real time. The combined positioning P-Box 102 is a device that combines GNSS (Global Satellite Navigation System) positioning, IMU (Inertial Navigation Positioning) and high-precision positioning RTK services to provide accurate position and attitude information.

[0093] The cockpit host 200 is used to interact with the outside world, perform visual display and receive trigger instructions; wherein the cockpit host 200 includes a world restoration module 201 and a vehicle following target status display module 202, wherein the world restoration module 201 is used for visual display, and through the virtual restoration of the scene 3D driving scene, the dynamic world around the vehicle is realistically presented, and the vehicles and road facilities are brightly colored and rich in details, and the lanes are distinguished and marked with different colors. The driver can use gestures to zoom, translate, and rotate the operation interface to observe the road conditions in all directions. The vehicle following target status display module 202 is used to display the range of vehicle-following targets and receive the driver's vehicle-following vehicle selection instructions. In one embodiment, the cockpit host 200 is a cockpit domain DHU (Digital Cockpit Head Unit).

[0094] The controller 300 is connected to the data acquisition module 100 and the cockpit host 200 in communication. The controller 300 is used to integrate the driving environment information and visualize it through the cockpit host 200, and generate a following strategy based on the trigger instruction fed back by the cockpit host 200, and generate corresponding control instructions based on the following strategy to automatically control the vehicle to perform acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle. In one embodiment, the controller 300 is an ADAS Domain Controller Unit (ADCU).

[0095] The controller 300 includes a perception fusion module 301 , a fusion positioning module 302 , a map data engine module 303 and an environment model module 304 .

[0096] Among them, the fusion positioning module 302 is used to integrate the driving environment information collected by the data acquisition module 100 to output the coordinates of the vehicle and its relative position in the lane; the fusion positioning module 302 combines the data measured by GNSS, IMU, and RTK in the combined positioning P-Box 102 with the wheel speed information and the visual lane line information, and outputs the precise absolute longitude and latitude coordinates of the vehicle, determines the position of the vehicle in the global coordinate system, and the relative position information such as the lateral offset of the vehicle in the lane, the distance from the center line of the lane or the left and right lane lines, etc.

[0097] The map data engine module 303 is used to match map-related information according to the coordinates of the vehicle and the navigation path, and output map information, real-time navigation path information and positioning information; the map data engine module 303 matches the link information of the high-precision map data according to the position of the vehicle in the global coordinate system and the SD navigation path, and outputs the lane-level map information, real-time navigation path information and lane-level positioning information 2 km ahead of the vehicle in real time.

[0098] The environment model module 304 is used to extract the map semantic level, road topology and navigation information, output the map lane lines, and construct a static global path; the environment model module 304 provides the decision-making planning module with lane change information so that the decision-making planning module can determine whether the lane change conditions are met.

[0099] The perception fusion module 301 is used to identify the target of the environmental information collected by the data acquisition module 100. The perception fusion module 301 performs time alignment and spatial coordinate conversion alignment on the information collected by the extravehicular sensor 101, and then performs target recognition on the image, identifies the target objects such as vehicles, pedestrians, traffic signs, obstacles in the image, and gives the position coordinates and category information in the image. Target recognition can optionally be performed by a deep learning algorithm CNN convolutional network. The target recognition result is combined with the point cloud information obtained by radar scanning to output the distance and angle data of the target object, and the speed, direction, coordinate position and category of the fused target object are output using a fusion method such as Kalman filtering. The image is semantically segmented by the lane line detection algorithm, and the lane line position, type, lane line divergence and merging point information, curb and other lane information output by the extravehicular sensor 101 are integrated, and the trend of the deviation between the map lane line and the visual lane line output by the environmental model module 304 is combined to perform real-time fusion correction.

[0100] The controller 300 also includes a decision-making planning module 305 , which is used to dynamically generate an optimal movement trajectory in real time based on the information of the perception fusion module 301 and the environment model module 304 .

[0101] The controller 300 further includes a control module 306 , and the control module 306 is configured to output a control signal according to the movement trajectory.

[0102] Actuator 400 includes a throttle, a steering wheel, and a braking system, and is used to receive control signals and adjust the speed and direction to achieve intelligent following motion of the vehicle.

[0103] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. An intelligent vehicle following method, applicable to high-speed driving scenarios, characterized in that: The following steps are involved: Acquire the vehicle's driving environment information in real time and display it visually, wherein the driving environment information includes lanes, vehicles on the lanes, traffic signs, and obstacle information; In the displayed driving environment information of the vehicle, identifying a driving area and a following vehicle; receiving a trigger instruction, and triggering the selected following vehicle as a target following vehicle; Automatically generate a following strategy based on the lane where the target following vehicle is located, the location information of vehicles on the lane, traffic signs and obstacle information; Based on the following vehicle strategy, corresponding control instructions are generated to automatically control the vehicle to execute acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle.

2. The intelligent vehicle following method according to claim 1, characterized in that: The dividing of the drivable area and the following vehicles according to the environmental information comprises: According to the location information of lanes, vehicles on lanes, traffic signs, and obstacles, the movable path of the vehicle is calculated through a path planning algorithm; the drivable area and the vehicles that can be followed are divided according to the movable path of the vehicle.

3. The intelligent vehicle following method according to claim 1, characterized in that: The following car strategy includes: A moving path for the self-vehicle to move behind the target vehicle in the drivable area is obtained according to the position of the target following vehicle relative to the self-vehicle, and the self-vehicle moves behind the target vehicle along the moving path and follows the target vehicle.

4. The intelligent vehicle following method according to claim 3, characterized in that: The following vehicle strategy includes: when the target following vehicle is the first vehicle in front of the own vehicle in the lane, adjusting the own vehicle speed to move to the rear of the target vehicle to follow the vehicle.

5. The intelligent vehicle following method according to claim 4, characterized in that: The following vehicle strategy includes: when the target following vehicle is the vehicle in front of the adjacent lane, judging whether there is safe lane change space at present and within a safe time according to the position and speed of the vehicles in the adjacent lane; if there is safe lane change space, the self-vehicle changes lanes and moves to the rear of the target lane to follow the vehicle.

6. The intelligent vehicle following method according to claim 4, characterized in that: The following vehicle strategy includes: when the target following vehicle is the Nth vehicle in front of the own vehicle in the lane, judging whether the overtaking condition is met, and if the overtaking condition is met, the own vehicle changes lanes to an adjacent lane, accelerates and changes lanes again to the rear of the target vehicle, wherein N is greater than 1.

7. An intelligent vehicle following system, characterized in that: A smart vehicle following method for implementing any one of claims 1 to 6, comprising: A data acquisition module, the data acquisition module is used to collect driving environment information; wherein the driving environment information includes lanes, vehicles on the lanes, traffic signs, and obstacle information; The cockpit host is used to interact with the outside world, perform visual display and receive trigger instructions; A controller is communicatively connected with the data acquisition module and the cockpit host, and is used for visually displaying the driving environment information after fusing it through the cockpit host, and generating a following strategy according to a trigger instruction fed back by the cockpit host, and generating corresponding control instructions based on the following strategy to automatically control the vehicle to execute acceleration and deceleration, lane change, and overtaking, so as to complete the following driving of the self-vehicle relative to the target following vehicle.

8. The intelligent vehicle following system according to claim 7, characterized in that: The controller includes a perception fusion module, a fusion positioning module, a map data engine module and an environmental model module. The fusion positioning module is used to fuse the driving environment information collected by the data acquisition module to output the coordinates of the vehicle and the relative position in the lane; the map data engine module is used to match the map related information according to the coordinates of the vehicle in combination with the navigation path, and output map information, real-time navigation path information and positioning information; the environmental model module is used to extract the map semantic level, road topology structure and navigation information, and construct a static global path; the perception fusion module is used to perform target recognition on the environmental information collected by the data acquisition module.

9. The intelligent vehicle following system according to claim 8, characterized in that: The controller also includes a decision-making planning module, which is used to dynamically generate a movement trajectory in real time according to information from the perception fusion module and the environment model module.

10. The intelligent vehicle following system according to claim 9, characterized in that: The controller further comprises a control module, and the control module is used for outputting a control instruction according to the movement trajectory.

Images