A system and method for automatically driving through a construction section

Abstract

The present invention relates to a system and method for automatic driving through a construction section, the system includes an environmental information collection module, a high-precision map, a positioning module, a forward road construction judgment module, a speed planning module, an environmental reconstruction module, a path planning module, a control module and a prompt takeover module; the method includes: judging the presence of a construction section ahead, decelerating in advance, judging the type of road occupation, path planning and path validity judgment. The present invention makes full use of the detection capability of the camera to identify the characteristic signs of the construction section, and performs path planning according to different scenarios, thereby improving the scene coherence of the automatic driving assistance system and filling the gap that the current automatic driving assistance system cannot pass through the construction section by actively changing lanes or searching for a pass area.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile intelligence, and specifically relates to a system and method for automatically driving through a construction section. Background Art

[0002] The rapid development of driving assistance technology has gradually become the standard configuration of mid-to-high-end models. One of the directions for the subsequent development of driving assistance systems is to improve the coverage of autonomous driving scenarios and the ability to handle abnormal working conditions. At present, advanced driving assistance systems can combine navigation path information to achieve point-to-point fully automatic driving on highways through autonomous lane changing and autonomous entry and exit ramps. Existing technologies such as: CN106708040B provides a sensor module, an autonomous driving system and a method for an autonomous driving system. The sensor module described in the patent can cover the vehicle driving environment in a full range, increase the environmental perception ability of the autonomous driving system, and effectively solve the problems of driving, lane changing and parking in autonomous driving, improve the reliability of the autonomous driving system, reduce safety risks, and ensure the safe operation of the autonomous driving system; there is also a vehicle roadblock avoidance and real-time planning of new routes system provided by CN107323455A, which adopts an infrared camera installed at the front bumper position, line identification and remote control update navigation, which can avoid temporary road closures and regularly update routes and road conditions according to the received roadblock information.

[0003] However, when dealing with road conditions on construction sections, the above-mentioned technical solutions and currently mass-produced driving assistance systems can generally only take over by issuing alarm prompts, and are unable to pass through the construction section by actively changing lanes or searching for a pass area. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a system and method for automatically driving through a construction section, so as to avoid the problem that the existing driving assistance system cannot pass through the construction section by actively changing lanes or searching for a pass area.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A system for automatically driving through a road construction section, comprising:

[0007] An environmental information acquisition module includes a high-resolution camera mounted on the current vehicle, the output information of which includes lane curve equations of the current lane and adjacent lanes, roadblock targets, and drivable area points ahead;

[0008] High-precision maps, which provide information on lane lines and road boundaries;

[0009] Positioning module, used in conjunction with high-precision maps, can provide centimeter-level positioning;

[0010] The road construction judgment module ahead is used to judge whether there is construction on the road ahead of the current vehicle based on the output information of the environment information collection module and the high-precision map;

[0011] The speed planning module is used to determine whether the current vehicle needs to be controlled to slow down according to the output information of the environmental information acquisition module. If deceleration is required, the deceleration rate is calculated to limit the speed;

[0012] The environment reconstruction module is used to reconstruct the environment based on the output information of the environment information acquisition module and the high-precision map to determine whether there is a complete drivable lane ahead;

[0013] A path planning module is used to plan a path according to the judgment result of the environment reconstruction module and to judge the validity of the planned path;

[0014] The control module is used to receive the final output result of the path planning module to control whether the vehicle passes;

[0015] The prompt takeover module is used to remind the driver to take over the vehicle when the current vehicle cannot plan a passable path.

[0016] The present invention also relates to a method for automatically driving through a road construction section. The method is based on the above-mentioned system for automatically driving through a road construction section and specifically includes the following steps:

[0017] S1: judging whether there is a construction section ahead. The road construction judgment module ahead judges whether there is road construction ahead according to the output information of the environment information collection module and the high-precision map. If there is, execute step S2;

[0018] S2: Slow down in advance. The speed planning module makes a speed limit plan based on the output result of the road construction judgment module ahead and the output information of the environment information collection module and the high-precision map;

[0019] S3: Determine the type of road occupation. When the vehicle approaches the construction section, the environment reconstruction module determines the lane traffic relationship based on the received lane and road condition information, thereby determining the path planning method to be used.

[0020] S4: path planning, according to the result of the road occupation type judgment obtained in step S3, the path planning module selects a corresponding path planning method, and the path planning method includes lane change path planning and passable path planning;

[0021] S5: Path validity judgment: the path planning module judges the validity of the generated path. If it is valid, the control module controls the vehicle to pass along the generated path. If it is invalid, the prompt takeover module prompts the driver to take over the vehicle.

[0022] To further improve the above technical solution, in step S1,

[0023] The high-precision map gives the distance between the current vehicle and the construction site and the lane occupied by the construction according to the current lane of the vehicle and the road ahead;

[0024] The environmental information collection module identifies that there are roadblocks or construction signs or speed limit signs ahead for closing the road, and the road construction judgment module ahead determines that there is a construction section ahead;

[0025] The roadblock comprises a cone-shaped barrel.

[0026] Furthermore, in step S2,

[0027] The environmental information collection module recognizes that there are construction signs and speed limit signs ahead, and the speed planning module sets the speed limit according to the speed limit information on the speed limit sign;

[0028] The environmental information collection module recognizes that there is a roadblock on the road ahead, and the speed planning module determines that the current lane is impassable. The starting point of the construction section is used as the braking position, and the deceleration is calculated according to the current vehicle speed to limit the speed. The speed limit state is not released until the path judgment is valid in step S5.

[0029] Furthermore, in step S3,

[0030] When slowing down and approaching the construction section, the environment reconstruction module determines the lane traffic relationship based on the position relationship between lane lines and roadblocks provided by the environment information collection module and the high-precision map;

[0031] If it is judged as one-way lane occupation, the unoccupied lane on the other side will be used as the target lane;

[0032] If it is judged that the road is occupied on both sides, subsequent drivable path planning is required.

[0033] Furthermore, in step S4, according to the result of determining the road occupation type obtained in step S3,

[0034] If it is a single-side lane occupation, the path planning module plans a fifth-order polynomial curve path based on the center position and heading of the target lane and the initial state of the current vehicle to complete the lane change path planning;

[0035] If the road is occupied on both sides, the path planning module will establish a grid map based on the lane lines and roadblock information from the environmental information acquisition module and the high-precision map output, and use the hybrid A* algorithm to search for a passable path to complete the passable path planning.

[0036] Furthermore, in step S5,

[0037] The path planning module needs to judge the validity of the results of lane change path planning or passable path planning, and determine whether the path planning result is controllable and followable and whether the current vehicle has a risk of collision with other targets; if the judgment result is that the control can be followed and there is no collision risk, it is deemed valid; if the judgment result is not controllable and followable, it is deemed invalid.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The system and method for automatically driving through a construction section of the present invention fully utilize the detection capability of a camera to identify characteristic signs of the construction section, and perform path planning according to different scenarios, thereby improving the scene coherence of the automatic driving assistance system and filling the gap that the current automatic driving assistance system cannot pass through the construction section by actively changing lanes or searching for a pass area. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flowchart of a method for automatically driving through a construction section according to an embodiment;

[0041] Figure 2 This is a schematic diagram of a one-side road occupation scenario;

[0042] Figure 3 This is a schematic diagram of the scene of double-sided road occupation. DETAILED DESCRIPTION

[0043] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0044] See also Figure 1 , a specific embodiment of a system for automatically driving through a construction section includes:

[0045] An environmental information acquisition module includes a high-resolution camera mounted on the current vehicle, the output information of which includes lane curve equations of the current lane and adjacent lanes, roadblock targets, and drivable area points ahead;

[0046] High-precision maps, which provide information on lane lines and road boundaries;

[0047] Positioning module, used in conjunction with high-precision maps, can provide centimeter-level positioning;

[0048] The road construction judgment module ahead is used to judge whether there is construction on the road ahead of the current vehicle based on the output information of the environment information collection module and the high-precision map;

[0049] The speed planning module is used to determine whether the current vehicle needs to be controlled to slow down according to the output information of the environmental information acquisition module. If deceleration is required, the deceleration rate is calculated to limit the speed;

[0050] The environment reconstruction module is used to reconstruct the environment based on the output information of the environment information acquisition module and the high-precision map to determine whether there is a complete drivable lane ahead;

[0051] A path planning module is used to plan a path according to the judgment result of the environment reconstruction module and to judge the validity of the planned path;

[0052] The control module is used to receive the final output result of the path planning module to control whether the vehicle passes;

[0053] The prompt takeover module is used to remind the driver to take over the vehicle when the current vehicle cannot plan a passable path.

[0054] A system for automatically driving through a construction section in an embodiment fully utilizes the detection capability of a camera to identify characteristic signs of the construction section and performs path planning according to different scenarios, thereby improving the scene coherence of the automatic driving assistance system and filling the gap that current automatic driving assistance systems are unable to pass through construction sections by actively changing lanes or searching for a pass area.

[0055] When implemented, the high-resolution camera in the environmental information acquisition module can identify the speed limit sign temporarily placed on the lane, and can detect small targets such as cones more than 150m away;

[0056] Please continue to participate Figure 1-Figure 3 The present invention also provides a method for automatically driving through a road construction section. The method is based on the above-mentioned system for automatically driving through a road construction section and specifically includes the following steps:

[0057] S1: judging whether there is a construction section ahead. The road construction judgment module ahead judges whether there is road construction ahead according to the output information of the environment information collection module and the high-precision map. If there is, execute step S2;

[0058] S2: Slow down in advance. The speed planning module makes a speed limit plan based on the output result of the road construction judgment module ahead and the output information of the environment information collection module and the high-precision map;

[0059] S3: Determine the type of road occupation. When the vehicle approaches the construction section, the environment reconstruction module determines the lane traffic relationship based on the received lane and road condition information, thereby determining the path planning method to be used.

[0060] S4: path planning, according to the result of the road occupation type judgment obtained in step S3, the path planning module selects a corresponding path planning method, and the path planning method includes lane change path planning and passable path planning;

[0061] S5: Path validity judgment: the path planning module judges the validity of the generated path. If it is valid, the control module controls the vehicle to pass along the generated path. If it is invalid, the prompt takeover module prompts the driver to take over the vehicle.

[0062] Wherein, in the step S1,

[0063] The high-precision map gives the distance between the current vehicle and the construction site and the lane occupied by the construction according to the current lane of the vehicle and the road ahead;

[0064] The environmental information collection module identifies that there are roadblocks or construction signs or speed limit signs ahead for closing the road, and the road construction judgment module ahead determines that there is a construction section ahead;

[0065] The roadblock comprises a cone-shaped barrel.

[0066] Wherein, in the step S2,

[0067] The environmental information collection module recognizes that there are construction signs and speed limit signs ahead, and the speed planning module sets the speed limit according to the speed limit information on the speed limit sign;

[0068] The environmental information collection module recognizes that there is a roadblock on the road ahead, and the speed planning module determines that the current lane is impassable. The starting point of the construction section is used as the braking position, and the deceleration is calculated according to the current vehicle speed to limit the speed. The speed limit state is not released until the path judgment is valid in step S5.

[0069] Wherein, in the step S3, when the vehicle is decelerating and approaching the construction section, the environment reconstruction module determines the lane traffic relationship according to the position relationship between the lane lines and roadblocks provided by the environment information collection module and the high-precision map;

[0070] If it is judged as one-way lane occupation, the unoccupied lane on the other side will be used as the target lane;

[0071] If it is judged that the road is occupied on both sides, subsequent drivable path planning is required.

[0072] Wherein, in said step S4, according to the result of judging the road occupation type obtained in said step S3,

[0073] If it is a single-side lane occupation, the path planning module plans a fifth-order polynomial curve path based on the center position and heading of the target lane and the initial state of the current vehicle to complete the lane change path planning;

[0074] If the road is occupied on both sides, the path planning module will establish a grid map based on the lane lines and roadblock information from the environmental information acquisition module and the high-precision map output, and use the hybrid A* algorithm to search for a passable path to complete the passable path planning.

[0075] When implementing, see Figure 2 and Figure 3 , Figure 2 The scenario shown is a one-way lane occupation. The front of the lane where the current vehicle is located is occupied by roadblock A, and the right lane is passable. The environment reconstruction module takes the right lane as the target lane, and then the path planning module plans a quintic polynomial curve path through polynomial fitting, that is, Figure 2 Curve B in the figure completes the lane change path planning; Figure 3 The scenario shown is a double-sided lane occupation. There are roadblocks A on both sides of the lane where the current vehicle is located. The passable area ahead is limited by the roadblocks on both sides rather than the lane lines. Therefore, under this condition, after the environment reconstruction module determines that the lane is occupied on both sides, the path planning module builds a grid map and uses the hybrid A* algorithm to search for a passable path ahead, that is, Figure 3 Curve C in .

[0076] Among them, in the step S5, the path planning module needs to judge the validity of the results of the lane change path planning or the passable path planning, and judge whether the path planning result is controllable to follow and whether the current vehicle has a collision risk with other targets; if the judgment result is that the control can be followed and there is no collision risk, it is deemed valid; if the judgment result is not that the control can be followed and there is no collision risk, it is deemed invalid.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A system for automatically driving through a construction section, characterized in that: include: An environmental information acquisition module includes a high-resolution camera mounted on the current vehicle, the output information of which includes lane curve equations of the current lane and adjacent lanes, roadblock targets, and drivable area points ahead; High-precision maps, which provide information on lane lines and road boundaries; Positioning module, used in conjunction with high-precision maps, can provide centimeter-level positioning; The road construction judgment module ahead is used to judge whether there is construction on the road ahead of the current vehicle based on the output information of the environment information collection module and the high-precision map; The speed planning module is used to determine whether the current vehicle needs to be controlled to slow down according to the output information of the environmental information acquisition module. If deceleration is required, the deceleration rate is calculated to limit the speed; The environment reconstruction module is used to reconstruct the environment based on the output information of the environment information acquisition module and the high-precision map to determine whether there is a complete drivable lane ahead; A path planning module is used to plan a path according to the judgment result of the environment reconstruction module and to judge the validity of the planned path; The control module is used to receive the final output result of the path planning module to control whether the vehicle passes; The takeover prompt module is used to remind the driver to take over the vehicle when the current vehicle cannot plan a passable path; When slowing down and approaching the construction section, the environment reconstruction module determines the lane traffic relationship based on the position relationship between lane lines and roadblocks provided by the environment information collection module and the high-precision map; If it is judged as one-way lane occupation, the unoccupied lane on the other side will be used as the target lane; If it is judged that the road is occupied on both sides, subsequent drivable path planning is required.

2. A method for automatically driving through a construction section, characterized in that: This method is based on the system for automatically driving through a road construction section as described in claim 1, and specifically includes the following steps: S1: judging whether there is a construction section ahead. The road construction judgment module ahead judges whether there is road construction ahead according to the output information of the environment information collection module and the high-precision map. If there is, execute step S2; S2: Slow down in advance. The speed planning module makes a speed limit plan based on the output result of the road construction judgment module ahead and the output information of the environment information collection module and the high-precision map; S3: Determine the type of road occupation. When the vehicle approaches the construction section, the environment reconstruction module determines the lane traffic relationship based on the received lane and road condition information, thereby determining the path planning method to be used. S4: path planning, according to the result of the road occupation type judgment obtained in step S3, the path planning module selects a corresponding path planning method, and the path planning method includes lane change path planning and passable path planning; S5: Path validity judgment: the path planning module judges the validity of the generated path. If it is valid, the control module controls the vehicle to pass along the generated path. If it is invalid, the prompt takeover module prompts the driver to take over the vehicle. In the step S3, When slowing down and approaching the construction section, the environment reconstruction module determines the lane traffic relationship based on the position relationship between lane lines and roadblocks provided by the environment information collection module and the high-precision map; If it is judged as one-way lane occupation, the unoccupied lane on the other side will be used as the target lane; If it is judged that the road is occupied on both sides, subsequent drivable path planning is required.

3. The method for automatically driving through a construction section according to claim 2, characterized in that: In the step S1, The high-precision map gives the distance between the current vehicle and the construction site and the lane occupied by the construction according to the current lane of the vehicle and the road ahead; The environmental information collection module identifies that there are roadblocks or construction signs or speed limit signs ahead for closing the road, and the road construction judgment module ahead determines that there is a construction section ahead; The roadblock comprises a cone-shaped barrel.

4. The method for automatically driving through a construction section according to claim 2, characterized in that: In the step S2, The environmental information collection module recognizes that there are construction signs and speed limit signs ahead, and the speed planning module sets the speed limit according to the speed limit information on the speed limit sign; The environmental information collection module recognizes that there is a roadblock on the road ahead, and the speed planning module determines that the current lane is impassable. The starting point of the construction section is used as the braking position, and the deceleration is calculated according to the current vehicle speed to limit the speed. The speed limit state is not released until the path judgment is valid in step S5.

5. The method for automatically driving through a construction section according to claim 2, characterized in that: In step S4, according to the result of the road occupation type determination obtained in step S3, If it is a single-side lane occupation, the path planning module plans a fifth-order polynomial curve path based on the center position and heading of the target lane and the initial state of the current vehicle to complete the lane change path planning; If the road is occupied on both sides, the path planning module will establish a grid map based on the lane lines and roadblock information from the environmental information acquisition module and the high-precision map output, and use the hybrid A* algorithm to search for a passable path to complete the passable path planning.

6. The method for automatically driving through a construction section according to claim 5, characterized in that: In the step S5, The path planning module needs to judge the validity of the results of lane change path planning or passable path planning, and determine whether the path planning result is controllable and followable and whether the current vehicle has a risk of collision with other targets; if the judgment result is that the control can be followed and there is no collision risk, it is deemed valid; if the judgment result is not controllable and followable, it is deemed invalid.

Images