A path control system and method for high-speed navigation and autonomous driving

Abstract

This invention discloses a path control system and method for high-speed navigation autonomous driving, including an onboard camera, an autonomous driving domain controller, and a cockpit controller. The onboard camera is used to collect image data around the vehicle, and its output is connected to the autonomous driving domain controller. The autonomous driving domain controller identifies the vehicle's current lane position based on the image data. The autonomous driving domain controller controls the vehicle's lane switching based on the destination information and current lane position from the navigation map on the cockpit controller. The advantage of this invention is that it can control and locate the vehicle's lane changes without relying on high-precision maps during high-speed navigation autonomous control, reducing costs.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving control paths for automobiles, and particularly to a path control system and method for high-speed navigation autonomous driving. Background Technology

[0002] Current high-speed navigation-based autonomous driving systems select driving routes by first accurately locating the vehicle using high-precision maps and positioning, then combining this with ordinary navigation maps to determine the driving path and subsequently control the vehicle, thereby enabling lane changes during highway autonomous driving. The core of this control lies in two parameters: high-precision positioning and high-precision maps. Deviations in either will lead to errors or defects in highway autonomous driving, affecting its reliability and accuracy.

[0003] In lane-level navigation-based autonomous driving control, the vehicle's current lane position on the highway is typically determined first. Then, based on the lane position and the actual needs of the navigation map, the autonomous driving domain controller issues a lane-change command. When a lane change is required, the vehicle is controlled to change lanes based on the position information of vehicles in front and behind, thus achieving reliable lane changing or lane control for autonomous driving on highways. However, all of this relies on high-precision positioning and high-precision maps. This method is easily affected by the update frequency of high-precision maps and is also costly, making it unsuitable for large-scale promotion. Furthermore, in some areas where high-precision positioning is inaccurate, it may also lead to the inability to perform lane-change control for autonomous driving on highways. When the current lane position of the vehicle cannot be obtained, lane-change control and other functions of autonomous driving cannot be realized. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a path control system and method for high-speed navigation and automatic driving. This system can control and locate the vehicle's route change without relying on high-precision maps during high-speed navigation and automatic control, thereby reducing costs.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: a path control system for high-speed navigation autonomous driving, including an on-board camera, an autonomous driving domain controller, and a cockpit controller; the on-board camera is used to collect image data around the vehicle, and its output terminal is connected to the autonomous driving domain controller; the autonomous driving domain controller identifies the current lane position of the vehicle based on the image data; the autonomous driving domain controller controls the lane path switching control of the vehicle according to the destination information of the navigation map of the cockpit controller and the current lane position.

[0006] The vehicle-mounted cameras include multiple cameras, which are installed in different locations on the vehicle.

[0007] The vehicle-mounted camera includes a front camera, a rear camera, and side cameras; wherein the front camera is located inside the windshield of the vehicle and is used to capture image data in front of the vehicle. The front camera includes a telephoto camera module and a conventional camera module. The telephoto camera module is used to capture distant targets in front, and the conventional camera module is used to capture images of targets, lane lines, and road markings around the vehicle at close and medium distances.

[0008] Side-view cameras are placed in pairs on the left and right sides of the vehicle to capture lane lines and objects in front of and behind the vehicle on both sides.

[0009] Rearview cameras are typically installed below the spoiler, inside the rear windshield, or at the tailgate switch to capture images of vehicles approaching from behind.

[0010] The autonomous driving domain controller includes a SOC and an MCU module, which are connected to the vehicle's camera via an LVDS line to acquire image data captured by the camera. The SOC chip stitches together images from different angles captured by the vehicle's camera to form a bird's-eye view. The stitched bird's-eye view includes road topology, lane lines, and objects, and obtains relevant information about the objects and lane lines based on the SOC's internal algorithm, such as object type, distance, speed, acceleration, heading angle, lateral distance of the vehicle from the left, right, left-left, and right-right lane lines, lane line curvature, and lane line curvature change rate. The MCU module determines the vehicle's current lane position and the number of highway lanes based on the object and lane line information provided by the SOC.

[0011] The cockpit controller is compatible with ordinary navigation maps and is used to provide navigation path information to the autonomous driving domain controller; the autonomous driving domain controller controls the vehicle's driving and lane changing based on the identified lane information of the current vehicle, as well as vehicle information and navigation information around the vehicle.

[0012] A control method for a path control system in high-speed navigation autonomous driving includes, after the autonomous driving function is activated during high-speed navigation, processing image data collected by a camera to obtain a stitched bird's-eye view, identifying road topology, lane lines, and objects in the stitched bird's-eye view to obtain the vehicle's current lane information, and then controlling the vehicle's driving and lane changing based on the identified lane information, vehicle information around the vehicle, and navigation information.

[0013] When the high-speed navigation autonomous driving function is activated, the system determines the vehicle's lane based on the road information in the bird's-eye view and records it as A. At the same time, it obtains the automatic control status of the vehicle from the autonomous driving domain controller in real time. When the autonomous driving domain controller controls the vehicle to automatically change lanes to the left or right according to the current road traffic conditions, it modifies and records the recorded lane A after each lane change. After a successful left lane change, the vehicle's lane becomes A+1, and after a successful right lane change, the vehicle's lane becomes A-1.

[0014] During high-speed autonomous driving, the bird's-eye view is updated in real time. If the right side of the road can be detected in the bird's-eye view, the accurate lane position B of the vehicle is determined based on the road edge and the lane position B is recorded. Then, the current lane position of the vehicle after lane change is recorded based on the lane position B. During the current high-speed navigation autonomous driving activation cycle, lane calculation is performed based on lane B. If the right side of the road is not detected in the current bird's-eye view, the vehicle continues to calculate the lane position based on lane A.

[0015] In high-speed autonomous driving mode, after receiving off-ramp information from the navigation map sent by the cockpit controller, the autonomous driving domain controller obtains the current lane information and then controls the vehicle to exit the off-ramp.

[0016] If the current lane position of the vehicle is calculated based on lane A, the autonomous driving domain controller will immediately control the vehicle to change lanes to the right after receiving the navigation reminder; if lane B is used as the reference, the autonomous driving domain controller will determine different right lane change timings based on the vehicle's current lane position and current traffic conditions to complete the driving path selection.

[0017] The advantages of this invention are: it enables vehicle lane change control and positioning without relying on high-precision maps during high-speed navigation automatic control, reducing costs. It locates the vehicle's position solely through BEV mapping, without depending on high-precision maps and positioning. With the support of ordinary navigation maps, it plans the driving path based on the vehicle's location. This method significantly reduces initial investment costs and facilitates large-scale deployment. Attached Figure Description

[0018] The following is a brief explanation of the contents of each of the accompanying drawings and the markings in the drawings:

[0019] Figure 1 This is a schematic diagram of the control system of the present invention;

[0020] Figure 2 This is a flowchart of the control method of the control system of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and the description of the preferred embodiments.

[0022] This solution obtains the vehicle's lane position through image recognition in the absence of high-precision maps and positioning, providing basic control parameters for high-speed navigation and autonomous driving control: lane position. It also records and controls lane position changes after lane changes, updating the current lane position in real time to meet the needs of high-speed autonomous driving. The solution includes:

[0023] (1) After the vehicle is powered on, it takes seven images of the front, rear, left and right sides of the vehicle through two front cameras, four side cameras and one rear camera, and inputs the seven images into the autonomous driving domain controller to stitch them together into a bird's-eye view.

[0024] (2) When the high-speed navigation automatic driving function is turned on, the bird's-eye view provides real-time road topology information, lane lines, vehicles, pedestrians, road signs, etc. The system evaluates the vehicle's lane position in the bird's-eye view and records the current lane number as A.

[0025] (3) During the lane change and overtaking process, the system records the lane position after the lane change. When the vehicle changes lanes from lane A to the left, the lane after the lane change is A+1; when the vehicle changes lanes from lane A to the right, the lane after the lane change is A-1. When the right curb can be identified in the bird's-eye view, the system calculates and records the accurate position B of the current lane of the vehicle.

[0026] (4) Upon receiving off-ramp information from a standard navigation map, the system controls the vehicle to gradually change lanes to the right. The system sets different right-change times based on the current lane position. That is, if the exact position of the current lane is determined, the system determines the right-change timing at different time gradients based on the current traffic flow and lane position; if the exact position of the current lane is not determined, the system immediately controls the vehicle to change lanes to the right upon receiving the above navigation information, until it changes to the rightmost lane, thus completing the selection of the driving path.

[0027] like Figure 1 The diagram shows the hardware system architecture of the path control system in high-speed navigation autonomous driving, including an onboard camera, an autonomous driving domain controller, and a cockpit controller. The onboard camera is used to collect image data around the vehicle, and its output is connected to the autonomous driving domain controller. The autonomous driving domain controller identifies the current lane position of the vehicle based on the image data. The autonomous driving domain controller controls the vehicle's lane path switching based on the destination information and current lane position of the navigation map from the cockpit controller.

[0028] The vehicle-mounted cameras consist of multiple units, each positioned in a different location within the vehicle. These include front cameras, rear cameras, and side cameras.

[0029] The front camera is located inside the windshield of the vehicle and is used to capture image data in front of the vehicle. The front camera includes a telephoto camera module and a conventional camera module. The telephoto camera module is used to capture distant targets in front, and the conventional camera module is used to capture images of targets, lane lines, and road markings around the vehicle at close and medium distances.

[0030] Side-view cameras are placed in pairs on the left and right sides of the vehicle to capture lane lines and objects in front of and behind the vehicle on both sides.

[0031] Rearview cameras are typically installed below the spoiler, inside the rear windshield, or at the tailgate switch to capture images of vehicles approaching from behind.

[0032] The autonomous driving domain controller includes a SOC and an MCU module, which are connected to the vehicle's camera via an LVDS line to acquire image data captured by the camera. The SOC chip stitches together images from different angles captured by the vehicle's camera to form a bird's-eye view. The stitched bird's-eye view includes road topology, lane lines, and objects, and obtains relevant information about the objects and lane lines based on the SOC's internal algorithm, such as object type, distance, speed, acceleration, heading angle, lateral distance of the vehicle from the left, right, left-left, and right-right lane lines, lane line curvature, and lane line curvature change rate. The MCU module determines the vehicle's current lane position and the number of highway lanes based on the object and lane line information provided by the SOC.

[0033] The cockpit controller is compatible with ordinary navigation maps and is used to provide navigation path information to the autonomous driving domain controller; the autonomous driving domain controller controls the vehicle's driving and lane changing based on the identified lane information of the current vehicle, as well as vehicle information and navigation information around the vehicle.

[0034] When the autonomous driving function is activated during high-speed navigation, the image data collected by the camera is processed to obtain a stitched bird's-eye view. Based on the stitched bird's-eye view, the road topology, lane lines, and objects in the image are identified to obtain the lane information where the vehicle is currently located. Then, based on the identified lane information, vehicle information around the vehicle, and navigation information, the vehicle's driving and lane changing are controlled.

[0035] like Figure 1 As shown, the present invention relates to two front cameras, four side-view cameras, one rear-view camera, one cockpit controller, and one autonomous driving domain controller.

[0036] Two front cameras are mounted on the inside of the vehicle's windshield to capture images of the area in front of the vehicle. One is a telephoto camera module used to capture distant targets, while the other is a standard camera module used to capture close- and medium-range targets, lane lines, road markings, etc.

[0037] Four side-view cameras are arranged in pairs on the left and right sides of the vehicle, usually located below the exterior rearview mirrors and on the B-pillar, to capture lane lines and objects on the left and right sides of the vehicle in front and behind.

[0038] A rearview camera is typically installed below the spoiler, inside the rear windshield, or at the tailgate switch to capture images of vehicles approaching from behind.

[0039] The cockpit controller is compatible with common navigation maps and can provide navigation route information to the autonomous driving domain controller.

[0040] The autonomous driving domain controller contains a SOC and an MCU module, which are connected to the seven camera modules mentioned above via LVDS cables. The SOC stitches together the seven video feeds into a bird's-eye view, which includes road topology, lane lines, and objects.

[0041] When the highway navigation autopilot function is activated, the system determines the vehicle's lane based on road information in the bird's-eye view and records it as lane A. Depending on the current traffic flow, the system controls the vehicle to automatically change lanes to the left or right. After changing lanes, the MCU records the vehicle's current lane position, following this procedure: a successful left lane change changes the vehicle's lane to A+1, and a successful right lane change changes the vehicle's lane to A-1.

[0042] The system constantly monitors the right-hand curb information. If the right-hand curb is detected in the bird's-eye view, the system determines the vehicle's accurate lane position B based on the curb and records it. During the current highway navigation autonomous driving activation cycle, lane calculation is performed based on lane B. If the right-hand curb is not detected in the current bird's-eye view, the vehicle continues to calculate its lane position based on lane A.

[0043] Upon receiving off-ramp information from a standard navigation map, if the vehicle is using lane A as a reference, the system will immediately control the vehicle to change lanes to the right after receiving the navigation prompt. If lane B is used as a reference, the system will determine different timings for changing lanes to the right based on the vehicle's current lane position and current traffic conditions: changing lanes earlier in the left lane and later in the right lane; changing lanes earlier when traffic is heavy and later when traffic is light. This strategy continues until the vehicle reaches the rightmost lane, completing the route selection. The lane position information obtained from the bird's-eye view is an estimated position because the rightmost curb may not be visible in the initial bird's-eye view. If it is not visible, the vehicle cannot determine which lane it is in. Therefore, lane A is an estimated lane; only when the rightmost curb is visible can the vehicle's exact lane position, i.e., lane B, be known. When you know your exact location B, you can determine when to change lanes based on the current traffic flow and your own location. However, if you don't know your exact location, for example, if there are 5 lanes in total, your car is on the far left, and you can only see 3 lanes from the bird's-eye view, then when you receive the navigation information to exit the lane, you must immediately change lanes to the right to prevent missing the ramp due to changing lanes too late.

[0044] The working principle of the high-speed navigation autonomous driving path selection based on high-precision maps described in this invention is as follows: After stitching together a bird's-eye view of 7 video feeds (front, rear, left, and right), the vehicle's lane position in the bird's-eye view is determined. If the right side of the road can be identified in the bird's-eye view, the accurate lane of the vehicle can be determined and recorded. Subsequent lane changes and exit strategies are based on this accurate lane. However, if the right side of the road is not identified throughout the entire function's activation period, the current lane A at the time the function is activated is used as the reference. When an exit lane warning is received, the vehicle immediately changes lanes to the right until it reaches the far right, thus completing the selection of the entire driving path within the function's activation period.

[0045] Obviously, the specific implementation of this invention is not limited to the above-described methods. Any non-substantial improvements made using the inventive concept and technical solution of this invention are within the protection scope of this invention.

Claims

1. A path control method for high-speed navigation and automatic driving, characterized in that: When the autonomous driving function is activated during high-speed navigation, the image data collected by the camera is processed to obtain a stitched bird's-eye view. Based on the stitched bird's-eye view, the road topology, lane lines, and objects in the image are identified to obtain the lane information where the vehicle is currently located. Then, based on the identified lane information, vehicle information around the vehicle, and navigation information, the vehicle's driving and lane changing are controlled. When the high-speed navigation autonomous driving function is activated, the system determines the vehicle's lane based on the road information in the bird's-eye view and records it as A; at the same time, it obtains the automatic control status of the vehicle from the autonomous driving domain controller in real time. When the autonomous driving domain controller controls the vehicle to automatically change lanes to the left or right according to the current road traffic conditions, it modifies and records the recorded lane A after each lane change. After a successful lane change to the left, the vehicle's lane becomes A+1, and after a successful lane change to the right, the vehicle's lane becomes A-1. During high-speed autonomous driving, the bird's-eye view is updated in real time. If the right side of the road edge can be detected in the bird's-eye view, the accurate lane position B of the vehicle is determined based on the road edge and the lane position B is recorded. Then, the current lane position of the vehicle after lane change is recorded based on the lane position B. During the current high-speed navigation autonomous driving activation cycle, lane calculation is performed based on lane B. If the right-side curb is not detected in the current bird's-eye view, the vehicle continues to calculate the lane position based on lane A. In high-speed autonomous driving mode, after receiving the off-ramp information from the navigation map sent by the cockpit controller, the autonomous driving domain controller obtains the current lane information and then controls the vehicle to exit the off-ramp. If the current lane position of the vehicle is calculated based on lane A, the autonomous driving domain controller will immediately control the vehicle to change lanes to the right after receiving the navigation reminder; if lane B is used as the reference, the autonomous driving domain controller will determine different right lane change timings based on the vehicle's current lane position and current traffic conditions to complete the driving path selection.

2. A path control system for high-speed navigation and automatic driving, characterized in that: The control system is used to run the control method as described in claim 1. The control system includes an on-board camera, an autonomous driving domain controller, and a cockpit controller. The on-board camera is used to collect image data around the vehicle, and its output is connected to the autonomous driving domain controller. The autonomous driving domain controller identifies the current lane position of the vehicle based on the image data. The autonomous driving domain controller controls the lane path switching of the vehicle according to the destination information of the navigation map of the cockpit controller and the current lane position.

3. The path control system for high-speed navigation and automatic driving as described in claim 2, characterized in that: The vehicle-mounted cameras include multiple cameras, which are installed in different locations on the vehicle.

4. The path control system for high-speed navigation and automatic driving as described in claim 3, characterized in that: The vehicle-mounted camera includes a front camera, a rear camera, and side cameras; wherein the front camera is located inside the windshield of the vehicle and is used to capture image data in front of the vehicle. The front camera includes a telephoto camera module and a conventional camera module. The telephoto camera module is used to capture distant targets in front, and the conventional camera module is used to capture images of targets, lane lines, and road markings around the vehicle at close and medium distances. Side-view cameras are placed in pairs on the left and right sides of the vehicle to capture lane lines and objects in front of and behind the vehicle on both sides. Rearview cameras are typically installed below the spoiler, inside the rear windshield, or at the tailgate switch to capture images of vehicles approaching from behind.

5. A path control system for high-speed navigation and automatic driving as described in claim 4, characterized in that: The autonomous driving domain controller includes a SOC and an MCU module, which are connected to the vehicle camera via an LVDS line to acquire image data captured by the camera. The SOC chip stitches together images from different angles captured by the vehicle camera to form a bird's-eye view. The stitched bird's-eye view includes road topology, lane lines, and objects, and obtains relevant information about objects and lane lines based on the SOC's internal algorithm. The MCU module determines the vehicle's current lane position and the number of highway lanes based on the object and lane line information provided by the SOC.

6. A path control system for high-speed navigation and automatic driving as described in any one of claims 2-5, characterized in that: The cockpit controller is compatible with ordinary navigation maps and is used to provide navigation path information to the autonomous driving domain controller; the autonomous driving domain controller controls the vehicle's driving and lane changing based on the identified lane information of the current vehicle, as well as vehicle information and navigation information around the vehicle.

Images