CN115009305B

CN115009305B - Narrow road passing processing method and narrow road passing processing device

Info

Publication number: CN115009305B
Application number: CN202210763299.0A
Authority: CN
Inventors: 李雪; 范圣印; 曲倩文
Original assignee: Beijing Yihang Yuanzhi Technology Co Ltd
Current assignee: Beijing Yihang Yuanzhi Technology Co Ltd
Priority date: 2022-06-29
Filing date: 2022-06-29
Publication date: 2023-04-11
Anticipated expiration: 2042-06-29
Also published as: CN116101325A; CN116101325B; CN115009305A

Abstract

The present disclosure provides a narrow road traffic processing method, including: judging whether a narrow road and/or a narrow road barrier exists in front of the vehicle based on the preset narrow road characteristics, and if so, starting a narrow road passing module of the vehicle; acquiring the pose of the vehicle at the starting time of the narrow road passing module; obtaining a panoramic top view of the host vehicle; detecting the obstacle vehicles based on the panoramic top view, and acquiring characteristic information of the obstacle vehicles, wherein the characteristic information comprises at least one of a front corner pose, a rear corner pose and a rearview mirror pose; calculating the shortest passable distance between the obstacle vehicle and the narrow road based on the characteristic information of the obstacle vehicle so as to preliminarily judge whether the vehicle can pass through without collision; preliminarily judging whether the vehicle can pass through without collision, and judging whether the vehicle can pass through without collision again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the characteristic width information of the vehicle; and when judging that the vehicle can pass through without collision again, planning the obstacle avoidance route.

Description

Narrow road passing processing method and narrow road passing processing device

Technical Field

The disclosure relates to the technical field of automatic driving, and relates to a narrow road passing processing method, a narrow road passing processing device, electronic equipment, a storage medium and an automobile.

Background

With the development of automotive technology, the driving of vehicles is facing more and more challenges, and the challenges especially refer to the passing of vehicles in narrow road sections, for example, when people pass in and out on roadside roads or in cells, the vehicles are almost all in a scene that one side of the road is parked with the vehicles, the other side is a road shoulder green belt or other obstacles, the space left in the middle can only accommodate one vehicle to pass through, sometimes the space is antagonistic, and even the rearview mirror needs to be folded up to be barely passed through. Or vehicles are illegally parked on one-way roads and road sides, the vehicles on the two sides of the roads are not parked orderly, the roads are narrow, and the vehicles are difficult to pass when the vehicles face each other. Or when the vehicle backs, the driver may scratch or scrape due to limitations of insufficient driving experience, insufficient visual field, narrow road and the like even if the attention is highly concentrated.

The existing backing collision avoidance device mainly utilizes radar ranging and other modes to sense the distance between a vehicle and an obstacle, the vehicle cannot use an ultrasonic radar to accurately position and identify the obstacle, and the obstacle cannot be visualized, so that a user cannot visually know the size, the position and the type of the obstacle. The cost of the mode is higher, and the structures of the reversing anti-collision device and the like in the prior art are complex.

Therefore, designing a method and a device for avoiding collision with other vehicles (other vehicle rearview mirrors and the like) during traffic and reversing under the condition of the narrow road has important theoretical significance and practical application value.

The following is a prior art solution.

The technical scheme 1: the Chinese patent application with publication number CN107097784A and invention name "a system and method for realizing low-speed automatic narrow road passing of an automobile" provides a system and method for realizing low-speed automatic narrow road passing of an automobile, wherein the system comprises an electronic control unit, a camera, an ultrasonic sensor, a touch display module, an electric power steering system, an electronic stabilizing system, an engine management system, an electronic gear shifting system and an electronic parking braking system, wherein the camera, the ultrasonic sensor, the touch display module, the electric power steering system, the electronic stabilizing system, the engine management system, the electronic gear shifting system and the electronic parking braking system are connected with the electronic control unit. The method comprises the following steps: after receiving a system activation request, the electronic control unit plans a driving path according to the information of obstacles around the vehicle when the vehicle meets the activation condition, and simultaneously determines a target gear, a target speed and a target rotation angle of the current driving; and then interacting with an electric power steering system, an electronic stabilizing system, an engine management system, an electronic gear shifting system and an electronic parking braking system, and keeping the vehicle to automatically run at a target speed and a target rotation angle according to a planned running path under a target gear. According to the technical scheme, the vehicle can automatically pass through a narrow road safely and conveniently, and collision accidents are avoided. However, in the technical scheme, a mode of jointly detecting by using the camera and the ultrasonic sensor is used when the obstacle information is determined, but error detection processing is not performed, and collision may occur due to inaccurate detection caused by weather and other reasons; meanwhile, objects occupying a small total amount of obstacles, such as vehicle mirrors, are easily ignored in general obstacle detection, resulting in erroneous judgment.

The technical scheme 2 is as follows: the Chinese patent application with the publication number of CN114043985A and the invention name of 'a narrow lane driving assistance method and device', provides a narrow lane driving assistance method and device, which relates to the technical field of assistant driving, the method comprises a narrow lane automatic driving process, and the narrow lane automatic driving process comprises the following steps: detecting the road width of a road in front of a vehicle; when narrow-lane automatic driving needs to be started, aligning the head and the body of the vehicle relative to a road in front of the vehicle at a speed not higher than a preset limited speed of narrow-lane automatic driving; and (5) retracting a rearview mirror of the vehicle, and controlling the vehicle to drive through a road in front of the vehicle at a speed not higher than the limited speed of automatic driving of the narrow road. This technical scheme carries out state discernment to road and vehicle self, and the control vehicle can assist the driver of different experience degrees according to predetermineeing speed through the lane, promotes the security that the vehicle drove, effectively ensures driver's personal safety, brings good driving experience for the driver. However, when the method identifies the vehicle obstacle, the width of the vehicle body is taken as a basis, and the rearview mirrors of the obstacle vehicle are not subjected to targeted processing, so that scratch is easy to occur; and in the actual traffic process, because of factors such as turning, the vehicle width is only ensured to be smaller than the road width, and the vehicle can not pass through without collision.

The technical scheme 3 is as follows: the Chinese patent application with the publication number of CN112141114A and the invention name of narrow passage auxiliary system and method provides a narrow passage auxiliary system and method, wherein the narrow passage auxiliary method comprises the steps of judging the passage feasibility, planning a driving path and controlling a vehicle to drive according to the planned driving path. The auxiliary method for narrow passage provided by the technical scheme comprises the following steps: 1. judging the traffic feasibility: acquiring a static image of a narrow lane, identifying narrow lane data information therefrom, and comparing the narrow lane data information with vehicle size information to determine whether the vehicle can pass through the narrow lane; if the vehicle is determined to be capable of passing through the narrow lane, entering step 2, and if the vehicle is determined to be incapable of passing through the narrow lane, informing a driver; 2. planning a driving path: dynamically acquiring images of narrow roads in the driving direction, calculating a central point of the narrowest road width as a track point of current vehicle driving, and planning a driving path by combining a distance detection result of an ultrasonic radar on surrounding obstacles; 3. controlling the vehicle to run according to the planned running path: and transverse and longitudinal control of the vehicle is completed according to the planned path, so that the vehicle is ensured to run on the planned path. The technical scheme improves the realizability of narrow-road passing in automatic driving, reduces the implementation cost and perfects the use scene of narrow-road passing. But the technical scheme does not carry out targeted treatment on the rearview mirror of the obstacle vehicle, and scratch is easy to occur; and in the actual traffic process, because of factors such as turning, the vehicle width is only ensured to be smaller than the road width, and the vehicle can not pass through without collision.

Technical scheme 1 has effectively reduced the vehicle collision accident that leads to because of the judgement mistake with the help of the interaction of a plurality of systems in the car, but technical scheme 1 does not carry out error check to data before the judgement, and has not handled like vehicle rear-view mirror etc. less but to judging accurate object that influences greatly to the accuracy to can lead to judging the problem that the rate of accuracy is difficult to promote.

Technical scheme 2 has combined discernment and adjusting device, effectively ensures driver's personal safety, but technical scheme 2 is based on the automobile body width when discerning the vehicle barrier, does not do the pertinence to the rear-view mirror of obstacle vehicle, and because factors such as turn in the actual traffic in-process, only guarantee that the car width is less than the road width and can not be certainly not can not have the collision and pass through, consequently in the actual application, to the narrow road that few parts can not have the collision and pass through probably produce the wrong judgement that can pass through.

Technical scheme 3 combines an auxiliary system and an auxiliary method, effectively improves the feasibility of narrow passage in automatic driving, reduces the implementation cost, and simultaneously perfects the use scene of narrow passage, but the technical scheme 3 does not specifically process the rearview mirrors of obstacle vehicles, vehicles with different obstacles and roads, and is still easy to scratch.

Disclosure of Invention

In order to solve at least one of the above technical problems, the present disclosure provides a narrow road passage processing method, a narrow road passage processing apparatus, an electronic device, a storage medium, and an automobile.

According to an aspect of the present disclosure, there is provided a narrow road passage processing method, including:

judging whether a narrow road and/or a narrow road obstacle exists in front of the vehicle based on the preset narrow road characteristics, and if so, starting a narrow road passing module of the vehicle;

acquiring the pose of the vehicle at the starting time of the narrow-path traffic module;

taking the starting time of the narrow-path passing module as a reference time, acquiring frame images of the surrounding environment of the vehicle, which are acquired by each camera of the vehicle in a preset time length, so as to obtain a panoramic top view of the vehicle;

detecting the barrier vehicle based on the panoramic top view to acquire characteristic information of the barrier vehicle, wherein the characteristic information comprises at least one of a front corner pose, a rear corner pose and a rearview mirror pose;

calculating the shortest passable distance between the obstacle vehicle and the narrow road based on the characteristic information of the obstacle vehicle so as to preliminarily judge whether the vehicle can pass through without collision;

if the vehicle is preliminarily judged to pass through without collision, judging whether the vehicle can pass through without collision again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information, wherein the vehicle characteristic width information comprises the width of the vehicle containing an external rearview mirror and/or the width of the vehicle not containing the external rearview mirror;

and when judging that the vehicle can pass through without collision again, planning the obstacle avoidance route.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, calculating the shortest passable distance between an obstacle vehicle and a narrow road based on feature information of the obstacle vehicle to preliminarily judge whether the own vehicle can pass through without collision includes:

whether the vehicle can pass through without collision is preliminarily judged based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety factor (xi).

According to the narrow-path passing processing method of at least one embodiment of the disclosure, whether a narrow path and/or a narrow-path obstacle exists in front of a vehicle is judged based on preset narrow-path characteristics, and if the narrow-path obstacle exists, a narrow-path passing module of the vehicle is started, and the method comprises the following steps:

acquiring a video image of a visual field in front of the vehicle in real time to judge narrow road and narrow road obstacles;

judging the distance between the vehicle and the narrow road and/or the narrow road obstacle when the narrow road and/or the narrow road obstacle exists;

and when the distance between the vehicle and the narrow road and/or the narrow road barrier is smaller than or equal to a preset threshold value, starting the narrow road passing module.

According to the narrow-road passage processing method of at least one embodiment of the present disclosure, acquiring the pose of the vehicle at the starting time of the narrow-road passage module includes:

acquiring the nearest satellite positioning data acquisition time before the starting time of the narrow-path passing module;

acquiring first IMU data and acquisition time thereof which are earlier than the acquisition time of the nearest satellite positioning data, wherein the acquisition time is used as starting time; acquiring first IMU data and acquisition time thereof, wherein the first IMU data is later than the starting time of the narrow road passing module, and the acquisition time is used as end point time;

acquiring the translation amount of the vehicle along the direction right in front of the vehicle, the translation amount along the direction right left of the vehicle and the translation amount along the direction right above the vehicle, namely attitude transformation amount, from the acquisition time of the nearest satellite positioning data to the starting time of the narrow road passing module;

and acquiring the accurate pose of the vehicle at the starting time of the narrow-path traffic module based on the pose transformation quantity and the pose of the vehicle at the nearest satellite positioning data acquisition time.

According to the narrow-path traffic processing method of at least one embodiment of the present disclosure, with the starting time of the narrow-path traffic module as a reference time, acquiring frame images of the surrounding environment of the host vehicle, which are acquired by respective cameras of the host vehicle for a preset time length, to obtain a panoramic top view of the host vehicle, includes:

carrying out distortion removal on the frame image acquired by each camera to obtain the frame image of each camera after distortion removal;

for each camera, the time corresponding to the first frame of image collected by the camera is used as the initial time, and the position and orientation of the image feature points of the rest frame of image are adjusted based on the IMU data and the wheel speed data of the camera;

spatially aligning all frame images of each camera after the time alignment to a world coordinate system by a camera coordinate system;

and synthesizing the frame images of each camera after the time alignment and the space alignment, and obtaining the panoramic top view by combining the pose of the vehicle at the starting time of the narrow-path traffic module.

According to the narrow road traffic processing method of at least one embodiment of the present disclosure, the method for detecting an obstacle vehicle based on the panoramic top view to obtain characteristic information of the obstacle vehicle, where the characteristic information includes at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, includes:

performing target detection based on a deep learning model on the panoramic top view to obtain 2D frames, car logo information and head directions of all barrier vehicles in the panoramic top view;

acquiring the number of grounding points of wheels of the obstacle vehicle, the pose of the grounding points in a world coordinate system and geometric parameters of a vehicle body based on a 2D frame of the obstacle vehicle;

when the number of the grounding points of the wheels of the obstacle vehicle is more than or equal to 2, acquiring the relative poses of the two wheels at one side of the obstacle vehicle close to the vehicle, and calling a corresponding vehicle model from a vehicle model library stored on a memory based on at least one of the relative poses, the geometric parameters of the vehicle body of the obstacle vehicle and the vehicle logo information;

and calling a corresponding vehicle model, placing the vehicle model on the position of the barrier vehicle in the panoramic top view based on the grounding point of the wheels of the barrier vehicle and the direction of the head of the barrier vehicle, and acquiring the front corner pose, the rear corner pose and/or the rearview mirror pose of the barrier vehicle based on the vehicle model.

performing target detection based on a deep learning model on the panoramic top view to obtain 2D frames, car logo information and head directions of all barrier cars in the panoramic top view;

when the number of the grounding points of the wheels of the obstacle vehicle is more than or equal to 2, acquiring the relative poses of two wheels at one side of the obstacle vehicle close to the vehicle, and calling a corresponding vehicle model from a vehicle model library stored on a memory on the basis of at least one of the relative poses, the geometric parameters of the vehicle body of the obstacle vehicle and the vehicle logo information;

and acquiring the front corner pose, the rear corner pose and/or the rearview mirror pose of the obstacle vehicle based on the SLAM module of the vehicle without acquiring the corresponding vehicle model.

and acquiring the front corner position posture, the rear corner position posture and/or the rear view mirror position posture of the obstacle vehicle based on the SLAM module of the vehicle without acquiring the vehicle logo information of the obstacle vehicle. According to the narrow road traffic processing method of at least one embodiment of the present disclosure, the method for detecting an obstacle vehicle based on the panoramic top view to obtain characteristic information of the obstacle vehicle, where the characteristic information includes at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, includes:

and when the number of the grounding points of the wheels of the obstacle vehicle is less than or equal to 1, acquiring the front corner pose, the rear corner pose and/or the rearview mirror pose of the obstacle vehicle based on the SLAM module of the vehicle.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, the method for calculating the shortest distance between an obstacle vehicle and a narrow road based on the characteristic information of the obstacle vehicle to preliminarily judge whether the vehicle can pass through without collision includes:

and if the narrow road is a straight narrow road, and the obstacle vehicle is positioned on a first side of the straight narrow road, acquiring the minimum value of the front angle and the rear angle of the obstacle vehicle and the distance between the rearview mirror and the edge line of a second side, opposite to the first side, of the straight narrow road as the shortest distance.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, when one or more of the intersection points of the perpendicular line made to the edge line of the straight narrow road by the front corner, the rear corner and the mirror to the edge line of the straight narrow road and the edge line does not fall on the edge line of the straight narrow road, the shortest distance is obtained based on the following steps:

marking the intersection points which do not fall on the edge lines;

taking the minimum value in the distance between the unidentified intersection point and the edge line as a first minimum value;

acquiring straight narrow road edge line end points around the obstacle vehicles, and calculating the shortest distance between the straight narrow road edge line end points and the obstacle vehicles;

if the perpendicular line segment corresponding to the shortest distance between the edge line end point of the straight narrow road and the obstacle vehicle can fall on the obstacle vehicle, reserving the shortest distance and taking the shortest distance as a second minimum value;

and comparing the first minimum value with the second minimum value, and taking the smaller value as the shortest distance between the obstacle vehicle and the narrow road.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, if a perpendicular line segment corresponding to the shortest distance between the edge line end point of the straight narrow road and the obstacle vehicle fails to fall on the obstacle vehicle, the following processing is performed:

and calculating the length of a connecting line between the end point of the edge line of the straight narrow road and the front corner, the rear corner and the rearview mirror of the obstacle vehicle, and taking the minimum value as the second minimum value.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, calculating the shortest distance between an obstacle vehicle and a narrow road based on feature information of the obstacle vehicle to preliminarily judge whether the own vehicle can pass through without collision includes:

and if the narrow road is a broken line narrow road, and the obstacle vehicle is positioned in a break point area of the broken line narrow road, respectively judging whether the vehicle can pass through without collision by taking the broken line narrow road as two intersected straight line narrow roads.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, the type of the narrow road is a broken line narrow road, the obstacle vehicle is located in a break point area of the broken line narrow road, and whether the vehicle can pass through without collision is respectively determined by using the broken line narrow road as two intersected straight line narrow roads, including:

respectively calculating the distance from the front corner and the rear corner of the obstacle vehicle and the distance from the protruding point of the rearview mirror to each section of the straight narrow road section;

acquiring the shortest distance between the barrier vehicle and each section of straight narrow road section;

calculating the shortest distance between the barrier vehicle and a narrow road break point;

and taking the minimum value of the shortest distance between the barrier vehicle and each section of straight narrow road section and the shortest distance between the barrier vehicle and the broken line narrow road as the shortest distance between the barrier vehicle and the broken line narrow road.

the type of the narrow road is a curve narrow road, and the barrier vehicles are located on the first side of the curve narrow road, the curve narrow road is evenly divided into a plurality of approximate straight road sections with preset lengths, so that the shortest distance between the barrier vehicles and each approximate straight road section is obtained, and the shortest distance between the barrier vehicles and the curve narrow road is further obtained.

the barrier vehicles are positioned on two sides of the narrow road, the shortest distance between each barrier vehicle on the first side of the narrow road and each barrier vehicle on the second side of the narrow road is obtained, and then the shortest distance between the barrier vehicles on the two sides of the narrow road, namely the shortest passable distance, is obtained.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, the type of the narrow road is judged based on the panoramic top view.

According to the narrow road passing processing method of at least one embodiment of the disclosure, whether the vehicle can pass through without collision is preliminarily judged based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety factor, and the method comprises the following steps:

obtaining the value of the product of the width of the vehicle and a preset safety factor (xi);

judging whether the shortest distance between the obstacle vehicle and the narrow road is larger than or equal to the product of the width of the vehicle without the external rearview mirror and a preset safety factor or not;

if so, preliminarily judging that no collision can pass; if not, the collision-free passing is determined to be impossible.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, whether collision-free passing is possible is judged again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information, and the method includes:

and judging whether the vehicle can pass through without collision or not based on the shortest passable distance between the obstacle vehicle and the narrow road, the characteristic width information of the vehicle and the type of the narrow road.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, whether collision-free passing can be performed is judged based on the shortest passable distance between an obstacle vehicle and a narrow road, the vehicle characteristic width information and the narrow road type, and the method includes:

when the narrow road type is a straight narrow road, judging whether the shortest distance between the obstacle vehicle and the narrow road is larger than or equal to the product value of the width of the vehicle containing the outside rearview mirror and a preset safety factor or not;

if so, judging that the vehicle can run along the straight narrow road and pass through the straight narrow road; if not, the vehicle is judged to be capable of folding the exterior mirror to drive along the straight narrow road and pass through the straight narrow road.

According to the narrow road passing processing method of at least one embodiment of the present disclosure, whether collision-free passing is possible or not is judged based on the shortest passable distance between the obstacle vehicle and the narrow road, the vehicle characteristic width information, and the narrow road type, and the method includes:

if the vehicle is positioned at the inner side of a curve compared with the obstacle vehicle, making a perpendicular line from the narrow-path broken-point position along the direction vertical to the axis of the obstacle vehicle, taking the width of the vehicle without an external rear view mirror/the width of the vehicle with the external rear view mirror as a position p on the perpendicular line, and respectively making line segments with the length of 1/2 of the vehicle length from the position p to the front and the rear vertical lines;

and judging whether the line segments intersect with the narrow road or not, and judging that the vehicle can run through the narrow road.

the narrow road type is a broken line narrow road, the obstacle vehicle is located at a break point position of the broken line narrow road, and if the vehicle is located on the outer side of a curve compared with the obstacle vehicle, the judgment is carried out based on the following judgment logic:

the distance between the rear corner of the obstacle vehicle and the first section of the broken line narrow road is a line segment s ₁ (ii) a Front corner point p of obstacle vehicle ₂ The distance between the point and the second segment of the narrow road of the broken line is a line segment s ₂ ；

When the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning left, the line segment s ₁ The right end point is taken as p ₁ Dot, by line segment s ₂ Left end endpoint as p ₂ Point; when the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning to the right, with line segment s ₁ Left end endpoint as p ₁ Dot, by line segment s ₂ The right end point is taken as p ₂ Point;

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

The method comprises the steps of obtaining the length L of a diagonal line of a rectangular frame which takes the length of a vehicle as a long side and does not contain the width of a rearview mirror as a short side ₁ (ii) a The length L of the diagonal line of a rectangular frame with the length of the vehicle as the long side and the width of the vehicle containing a rearview mirror as the short side is obtained ₂ (ii) a And

comparison S _L And L ₁ 、L ₂ If s is the value of _n Is > ξ · b and s _L ＞ξ·L ₂ If so, judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi. B or xi. L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi a or s is less than or equal to _L ≤ξ·L ₁ Then the vehicle cannot pass through without collision.

the narrow road type is a curved narrow road, and more than two obstacle vehicles are positioned on one side of the curved narrow road; the decision is made based on the following decision logic:

obtaining the shortest distance line segment s between the first barrier vehicle and the curve narrow road ₁ Obtaining the shortest distance line segment s between the second barrier vehicle and the curve narrow road ₂ ；

When the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning left, the line segment s ₁ Right end point is taken as p ₁ Dot, by line segment s ₂ Left end endpoint as p ₂ Point; when the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning to the right, with line segment s ₁ Left end endpoint as p ₁ Dot, by line segment s ₂ The right end point is taken as p ₂ Point;

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

The diagonal length L of a rectangular frame with the length of the vehicle as the long side and the width of the vehicle without a rearview mirror as the short side is obtained ₁ (ii) a The length L of the diagonal line of a rectangular frame with the length of the vehicle as the long side and the width of the vehicle containing a rearview mirror as the short side is obtained ₂ (ii) a And

comparison S _L And L ₁ 、L ₂ If s is the value of _n B and s > xi _L ＞ξ·L ₂ Judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi. B or xi. L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi.a or s is less than or equal to _L ≤ξ·L ₁ Then the vehicle cannot pass through without collision.

for various narrow road types, obstacle vehicles are arranged on both sides of the narrow road, and more than two obstacle vehicles are arranged on one side of the narrow road, the judgment is carried out based on the following judgment logics:

obtaining the shortest distance line segment s between the first barrier vehicle and the second barrier vehicle ₁ And acquiring the shortest distance line segment s between the third barrier vehicle and the second barrier vehicle ₂ The first barrier vehicle and the third barrier vehicle are located on the same side of the narrow road, and the second barrier vehicle is located on the other side of the narrow road;

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

comparison S _L And L ₁ 、L ₂ If s is the value of _n Is > ξ · b and s _L ＞ξ·L ₂ Judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi. B or xi. L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi.a or s is less than or equal to _L ≤ξ·L ₁ Then the vehicle cannot pass through without collision.

According to the narrow-road passing processing method of at least one embodiment of the disclosure, when it is judged again that the vehicle can pass through without collision, obstacle avoidance route planning is performed, which includes:

and planning an obstacle avoidance route based on the shortest passable distance.

According to another aspect of the present disclosure, there is provided a narrow passage processing apparatus including:

the rough measurement module judges whether a narrow road and/or a narrow road obstacle exists in front of the vehicle or not based on preset narrow road characteristics, and if so, the narrow road passing module of the vehicle is started;

a narrow passage module activated in response to the rough measurement module;

the pose acquisition module acquires the pose of the vehicle at the starting time of the narrow road passing module;

the panoramic top view acquisition module is used for acquiring frame images of the surrounding environment of the vehicle, which are acquired by each camera of the vehicle in a preset time length, by taking the starting time of the narrow-path passing module as a reference time so as to acquire the panoramic top view of the vehicle;

the obstacle detection module is used for detecting an obstacle vehicle based on the panoramic top view and acquiring characteristic information of the obstacle vehicle, wherein the characteristic information comprises at least one of a front corner pose, a rear corner pose and a rearview mirror pose;

the narrow passage module comprises:

the first judgment module calculates the shortest passable distance between the obstacle vehicle and the narrow road based on the characteristic information of the obstacle vehicle so as to preliminarily judge whether the vehicle can pass through without collision;

a second judging module, configured to, if the first judging module preliminarily judges that the host vehicle can pass through without collision, judge again whether the host vehicle can pass through without collision based on at least a shortest passable distance between the obstacle vehicle and a narrow road and host vehicle characteristic width information, where the host vehicle characteristic width information includes a width that the host vehicle includes an external rearview mirror and/or a width that the host vehicle does not include an external rearview mirror;

and the route planning module is used for planning the obstacle avoidance route when the second judging module judges that the vehicle can pass through without collision again.

According to the narrow passage processing device of at least one embodiment of the present disclosure, calculating the shortest passable distance between an obstacle vehicle and a narrow passage based on feature information of the obstacle vehicle to preliminarily judge whether the own vehicle can pass through without collision includes:

whether the vehicle can pass through without collision is preliminarily judged based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety coefficient (xi).

According to the narrow passage processing apparatus of at least one embodiment of the present disclosure, the obstacle detection module is included in the narrow passage module.

According to the narrow passage processing apparatus of at least one embodiment of the present disclosure, the panoramic top view acquisition module and the obstacle detection module are included in the narrow passage module.

According to still another aspect of the present disclosure, there is provided an electronic device including: a memory storing execution instructions; and a processor executing the execution instructions stored by the memory, so that the processor executes the narrow-path passage processing method of any embodiment of the disclosure.

According to still another aspect of the present disclosure, a readable storage medium is provided, and the readable storage medium stores therein an execution instruction, which is executed by a processor to implement the narrow road passage processing method according to any one of the embodiments of the present disclosure.

According to yet another aspect of the present disclosure, there is provided an automobile including: the device comprises a vehicle-mounted camera device, a display screen, a wheel speed meter, an inertia measuring device, a processor and a readable storage medium; the readable storage medium is characterized by storing execution instructions, and the execution instructions are used for realizing the narrow-path traffic processing method of any embodiment of the disclosure when being executed by a processor.

According to yet another aspect of the present disclosure, there is provided an automobile including: the device comprises a vehicle-mounted camera device, a display screen, a wheel speed meter, an inertia measuring device, a processor and a readable storage medium; the narrow-path traffic processing device is characterized by further comprising a computer program, wherein the computer program is stored in the readable storage medium by executing the instruction module.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the disclosure and together with the description serve to explain the principles of the disclosure.

Fig. 1 is a schematic flow chart of a narrow passage processing method according to an embodiment of the present disclosure.

Fig. 2 is a time reference axis.

Fig. 3 is a schematic view of a vehicle with an obstacle on one side of a straight narrow road.

Fig. 4 is a schematic view of a vehicle with an obstacle on one side of a narrow road of a broken line (located in a broken point area).

Fig. 5 is a schematic view of a vehicle with an obstacle on one side of a curved narrow road.

Fig. 6 is a schematic view of a vehicle with obstacles on both sides of a narrow road.

Fig. 7 is a schematic view of the case of a multi-obstacle vehicle.

Fig. 8 is a schematic diagram of a case where the obstacle vehicle is located at the position of a break point and the host vehicle is located inside a curve.

Fig. 9 is a schematic diagram of a case where the obstacle vehicle is located at the position of a break point and the host vehicle is located outside the curve.

Fig. 10 is a schematic view of the situation in a curved narrow road.

Fig. 11 is a schematic view of a case where there is an obstacle vehicle on both sides.

Fig. 12 is a schematic diagram of obstacle avoidance route planning.

Fig. 13 is a specific flowchart of a narrow passage processing method according to an embodiment of the present disclosure.

Fig. 14 is a block diagram schematically illustrating a configuration of a narrow-passage processing apparatus implemented by hardware of a processing system according to an embodiment of the present disclosure.

Fig. 15 is a block diagram schematically illustrating the structure of a narrow-passage processing apparatus implemented by hardware using a processing system according to an embodiment of the present disclosure.

Detailed Description

The present disclosure will be described in further detail with reference to the drawings and embodiments. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant matter and not restrictive of the disclosure. It should be further noted that, for the convenience of description, only the portions relevant to the present disclosure are shown in the drawings.

It should be noted that the embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict. Technical solutions of the present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Unless otherwise indicated, the illustrated exemplary embodiments/examples are to be understood as providing exemplary features of various details of some ways in which the technical concepts of the present disclosure may be practiced. Accordingly, unless otherwise indicated, features of the various embodiments may be additionally combined, separated, interchanged, and/or rearranged without departing from the technical concept of the present disclosure.

The use of cross-hatching and/or shading in the drawings is generally used to clarify the boundaries between adjacent components. As such, unless otherwise noted, the presence or absence of cross-hatching or shading does not convey or indicate any preference or requirement for a particular material, material property, size, proportion, commonality between the illustrated components and/or any other characteristic, attribute, property, etc., of a component. Further, in the drawings, the size and relative sizes of components may be exaggerated for clarity and/or descriptive purposes. While example embodiments may be practiced differently, the specific process sequence may be performed in a different order than that described. For example, two processes described consecutively may be performed substantially simultaneously or in reverse order to that described. In addition, like reference numerals denote like parts.

When an element is referred to as being "on" or "on," "connected to" or "coupled to" another element, it can be directly on, connected or coupled to the other element or intervening elements may be present. However, when an element is referred to as being "directly on," "directly connected to" or "directly coupled to" another element, there are no intervening elements present. For purposes of this disclosure, the term "connected" may refer to physically, electrically, etc., and may or may not have intermediate components.

The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Furthermore, when the terms "comprises" and/or "comprising" and variations thereof are used in this specification, the stated features, integers, steps, operations, elements, components and/or groups thereof are stated to be present but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof. It is also noted that, as used herein, the terms "substantially," "about," and other similar terms are used as approximate terms and not as degree terms, and as such, are used to interpret inherent deviations in measured values, calculated values, and/or provided values that would be recognized by one of ordinary skill in the art.

The following describes the narrow passage processing method and the narrow passage processing apparatus of the present disclosure in detail with reference to fig. 1 to 15.

Fig. 1 is a flowchart illustrating a narrow passage processing method S100 according to an embodiment of the present disclosure.

Referring to fig. 1, a narrow passage processing method S100 of the present embodiment includes:

s102, judging whether a narrow road and/or a narrow road obstacle (a vehicle and the like) exists in front of the vehicle (rough measurement module 1002) based on preset narrow road characteristics, and if so, starting a narrow road passing module 1004 of the vehicle;

s104, acquiring the starting time (t) of the vehicle in the narrow road passing module 1004 _on ) The pose of (a);

s106, taking the starting time of the narrow-path passing module 1004 as a reference time, acquiring frame images of the surrounding environment of the vehicle, which are acquired by each camera of the vehicle within a preset time length (for example, 2S), so as to obtain a panoramic top view of the vehicle;

s108, detecting the barrier vehicle based on the panoramic top view, and acquiring characteristic information of the barrier vehicle, wherein the characteristic information comprises at least one of a front corner pose, a rear corner pose and a rearview mirror pose;

s110, calculating the shortest passable distance between the obstacle vehicle and the narrow road (namely the narrowest length of the vehicle needing to pass through in the narrow road passing process) based on the characteristic information of the obstacle vehicle so as to preliminarily judge whether the vehicle can pass through without collision;

s112, primarily judging whether the vehicle can pass through without collision, and then judging whether the vehicle can pass through without collision again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information, wherein the vehicle characteristic width information comprises the width of the vehicle including an external rearview mirror and/or the width of the vehicle not including the external rearview mirror;

and S114, when judging that the vehicle can pass through without collision again, planning an obstacle avoidance route.

According to a preferred embodiment of the present disclosure, calculating the shortest passable distance between an obstacle vehicle and a narrow road based on characteristic information of the obstacle vehicle to preliminarily determine whether the host vehicle can pass through without collision includes:

According to the narrow road passing processing method S100, S102, (rough measurement module 1002) of the preferred embodiment of the present disclosure, whether a narrow road and/or a narrow road obstacle (vehicle, etc.) exists in front of the vehicle is judged based on preset narrow road characteristics, if yes, the narrow road passing module 1004 of the vehicle is started, and the method comprises the following steps:

s1021, acquiring video images (namely a continuous static image sequence) of the front view of the vehicle in real time to judge narrow road and narrow road obstacles;

s1022, when the narrow road and/or the narrow road obstacle exists, judging the distance between the vehicle and the narrow road and/or the narrow road obstacle;

s1023, when the distance between the vehicle and the narrow road and/or the narrow road obstacle is less than or equal to a preset threshold (i.e. a safety threshold, for example, 3 meters), the narrow road passing module 1004 is started.

In the present disclosure, preferably, the vehicle coordinate system uses the center of the rear axle of the vehicle as the origin, the forward direction of the vehicle is the front, the x-axis of the vehicle coordinate system points to the front, the y-axis points to the left, and the z-axis points to the top; the camera coordinate system takes the optical center of the camera as an origin, the z axis of the camera coordinate system points to the front of the camera, the x axis points to the right of the camera, and the y axis points to the lower part of the camera; the laser radar coordinate system takes the geometric center of the radar as an origin, the z axis of the laser radar coordinate system points to the front of the radar, the x axis points to the right of the radar, and the y axis points to the lower part of the radar; the IMU (inertial measurement unit) coordinate system is consistent with the vehicle body coordinate system, the center of the rear shaft of the vehicle is taken as the origin, the advancing direction of the vehicle is the dead ahead, the x-axis points to the dead ahead, the y-axis points to the dead left, and the z-axis points to the dead above.

During the driving of the vehicle, the rough measurement module 1002 is always started. The rough measurement module 1002 may determine the narrow road and the narrow road obstacle by using a 3D object detection algorithm, a semantic segmentation algorithm, and the like in the prior art when finding that there is a narrow road and an obstacle vehicle in the driving area, and the present disclosure exemplarily sets 3 meters as a safety threshold, and when the distance between the host vehicle and the narrow road and the obstacle vehicle is less than or equal to the safety threshold, the narrow road passing module 1004 of the host vehicle is automatically started.

In the present disclosure, the host vehicle is preferably composed of a display screen, a processor, a memory, and six cameras (for example, fisheye cameras) respectively installed in front of, behind the left, in front of, behind the host vehicle, the six cameras simultaneously capturing images (video images) around the host vehicle, the display screen, the processor, and the memory being installed inside the host vehicle.

Both rough test module 1002 and narrow pass module 1004 described above in this disclosure are preferably software modules stored on a memory to be executable by a processor.

The narrow-road passing processing method S100, S104 according to the preferred embodiment of the present disclosure acquires the starting time (t) of the vehicle in the narrow-road passing module 1004 _on ) Includes:

s1041, obtaining the starting time (t) of the narrow road passing module 1004 _on ) Previous nearest neighbor satellite positioning data acquisition time (t) _GPS )；

S1042, acquiring the positioning data earlier than the nearest satellite positioning data acquisition time (t) _GPS ) And its acquisition time as a starting time (t) _start ) (ii) a Obtaining a start time (t) later than the narrow road passing module 1004 _on ) And its collection time as an end time (t) _end )；

S1043, acquiring the nearest satellite positioning data acquisition time (t) _GPS ) To the starting time (t) of the narrow road passing module 1004 _on ) The amount of translation (Δ x) of the host vehicle in the direction directly ahead of the host vehicle (x direction), the amount of translation (Δ y) in the direction directly to the left of the host vehicle, and the amount of translation (Δ z) in the direction directly above the host vehicle, that is, the attitude change amount (Δ T) _GPS-on )；

S1044 based on pose transformation quantity (delta T) _GPS-on ) And the nearest satellite positioning data acquisition time (t) _GPS ) Position and posture of the host vehicle (P) _GPS ) Acquiring the accurate pose (P) of the vehicle at the starting moment of the narrow-path traffic module 1004 _on )。

Upon activation of the narrow-path passage module 1004, the host vehicle acquires an approximate pose (P) of the host vehicle's vehicle, preferably based on a satellite-based positioning sensor _GPS ) Such as a Beidou sensor, a GPS sensor, etc. Since the satellite positioning sensor has a fixed acquisition frequency, it is not necessarily able to exactly acquire the pose of the vehicle at the start time of the narrow road traffic module 1004, and therefore, the present disclosure preferably accurately acquires the pose information of the vehicle by means of IMU data and wheel speed data.

In some embodiments of the present disclosure, the wheel speed data is used to obtain the speed of the host vehiclet _GPS To t _on The translation amount delta x along the x-axis direction in the time period; obtaining the t of the vehicle by using a linear interpolation method by means of IMU data _GPS To t _on The angles of rotation Δ yaw, Δ pitch and Δ roll in different directions x, y, z over a time period. The integral of the rotation angles delta yaw, delta pitch, delta roll and the translation amount delta x is used for obtaining the t-th of the vehicle _GPS To t _on The translation amounts delta y and delta z along the y-axis direction and the z-axis direction in a time period can obtain the t-axis translation amount of the vehicle _GPS To t _on Pose transformation quantity delta T in time period _GPS-on . According to P _G ＝ΔT _GPS-on P _o _n The accurate pose P of the vehicle at the starting time of the narrow passage module 1004 based on the coordinate system of the satellite positioning system (the GPS coordinate system is taken as an example in the disclosure) can be obtained _on And at the current pose P _on As the origin of the world coordinate system in the next narrow road pass.

Fig. 2 shows a time reference axis to more intuitively express the relationship between the above-mentioned respective moments.

The narrow-path traffic processing method S100, S106 according to the preferred embodiment of the present disclosure, which takes the starting time of the narrow-path traffic module 1004 as a reference time, acquires frame images of the surrounding environment of the host vehicle captured by each camera of the host vehicle for a preset time length (for example, 2S) to obtain a panoramic top view of the host vehicle, includes:

s1061, carrying out distortion removal on the frame image collected by each camera to obtain the frame image of each camera after distortion removal; the distortion removal of the frame image may adopt a distortion removal algorithm in the prior art, which is not particularly limited in the present disclosure;

s1062, regarding the moment (t) corresponding to the first frame image acquired by each camera as an initial moment, and performing image feature point posture adjustment (the image feature point can be an obstacle vehicle feature point) on the other frame images based on the vehicle IMU data and the wheel speed data so as to align the images to the initial moment (t) in time;

s1063, aligning all the frame images of each camera after time alignment to a world coordinate system by a camera coordinate system space;

s1064, synthesizing the time-aligned and spatially-aligned frame images of the cameras, and combining the start time (t) of the vehicle in the narrow-passage module 1004 _on ) The panoramic top view is obtained by the pose of the user.

The moment when each camera acquires the first frame image in the preset time length is the starting moment of the narrow-road passing module 1004.

In some embodiments of the present disclosure, the cameras of the host vehicle are classified into 6 cameras for front view, rear view, front left view, rear left view, front right view, and rear right view, depending on the installation location.

Its associated external reference is represented as a matrix T _CW The matrix T _CW Can be decomposed into a rotation matrix R _CW And a translation matrix t _CW They determine the relative positional relationship between the camera coordinates and the world coordinate system. Camera coordinates P _c With world coordinate P _w The relationship between the two is as follows: p _c ＝T _CW P _w I.e. P _c ＝R _CW P _w +t _CW 。

Its associated reference is represented by a matrix K comprising { f, κ, S } _x ,S _y ,C _x ,C _y Six parameters, where f is the focal length; kappa represents the magnitude of radial distortion, if the kappa is a negative value, the distortion is barrel-type distortion, and if the kappa is a positive value, the distortion is pillow-type distortion; s _x ,S _y Is a scaling factor representing the distance between adjacent pixels on the camera sensor in the horizontal and vertical directions; c _x ,C _y Is the perpendicular projection of the center of projection onto the imaging plane and is also the center of radial distortion.

In the present disclosure, the distortion of the cameras is radial distortion caused by the special shape of the lens, and data collected by six cameras is subjected to radial distortion removal processing. The radial distortion removal processing is performed, for example, using the following equation:

X _rcorr ＝X _p (1+k ₁ r ² +k ₂ r ⁴ +k ₃ r ⁶ )；Y _rcorr ＝Y _p (1+k ₁ r ² +k ₂ r ⁴ +k ₃ r ⁶ )

obtaining pose P of the feature point of the obstacle (such as the feature point of the obstacle vehicle) after distortion removal _c,rcorr 。

According to P _c ＝T _CW P _w Converting the pose of the characteristic point of the obstacle into P under the world coordinate system _w,rcorr . And (3) synthesizing the processed data of the 6 cameras (the data which can be observed by a plurality of cameras are superposed, and the data which can be observed by only one camera is spliced), finally forming a 360-degree panoramic top view without a gap around the vehicle, and projecting the top view into the BEV.

In order to enhance the applicability and the accuracy of obstacle vehicle detection, the information of obstacle vehicles is perfected by a dynamic multi-frame superposition mode based on the content of cameras, namely all the camera detection data are projected under a world coordinate system by taking the time of starting a narrow-path traffic module as a reference, so that enough data are supported in the process of obstacle vehicle detection.

In the present disclosure, the image captured by the camera is taken within a preset time period (e.g., 2s, which is adjustable). And taking the time t corresponding to the first frame image as an initial time. Adjusting the pose of the obstacle feature points of all the rest frame images based on the IMU data and the wheel speed data of the vehicle by the method in the step S104, and aligning the time to the time t; by means of P _c ＝T _CW P _w Frame images acquired by each camera within a preset time length are spatially aligned to a world coordinate system, the aligned camera data are accumulated, and the accumulated data are overlapped and spliced to obtain a panoramic top view and projected to a BEV (bird's eye view coordinate system).

According to the narrow road traffic processing method S100, S108 of the preferred embodiment of the present disclosure, the method detects an obstacle vehicle based on the panoramic top view, and obtains characteristic information of the obstacle vehicle, where the characteristic information includes at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, and includes:

s1081, performing target detection based on a deep learning model on the panoramic top view, and acquiring 2D frames, vehicle logo information and vehicle head directions of all obstacle vehicles in the panoramic top view;

s1082, acquiring the number of grounding points of wheels of the obstacle vehicle and the pose (P) of the grounding points in a world coordinate system based on the 2D frame of the obstacle vehicle _tyre,i ) And body geometry parameters (including body length, body width, and body diagonal length); where i =1,2, \8230;, N. N is less than or equal to 4, representing the number of detected tyres;

s1083, when the number of the grounding points of the wheels of the obstacle vehicle is greater than or equal to 2, acquiring the relative poses (delta P) of the two wheels at one side of the obstacle vehicle close to the vehicle, and calling a corresponding vehicle model from a vehicle model library 1012 stored in a memory on the basis of at least one of the relative poses, the vehicle body geometric parameters of the obstacle vehicle and the vehicle logo information;

s1084, calling a corresponding vehicle model, placing the vehicle model on the position of the obstacle vehicle in the panoramic top view based on the grounding point of the wheels of the obstacle vehicle and the direction of the vehicle head, and obtaining the position of the vehicle front corner, the position of the vehicle rear corner and/or the position of the vehicle rearview mirror of the obstacle vehicle based on the vehicle model.

s1082, acquiring the number of grounding points of wheels of the obstacle vehicle and the pose (P) of the grounding points in a world coordinate system based on the 2D frame of the obstacle vehicle _tyre,i ) And body geometry parameters (including body length, body width, and body diagonal length); wherein i =1,2, \8230, N; n is less than or equal to 4, representing the number of detected tyres;

and S1084, obtaining the front corner pose, the rear corner pose and/or the rearview mirror pose of the obstacle vehicle based on the SLAM module of the vehicle without obtaining the corresponding vehicle model.

According to the narrow road traffic processing method S100, S108 of the preferred embodiment of the present disclosure, the method detects the obstacle vehicle based on the panoramic top view, and obtains the characteristic information of the obstacle vehicle, where the characteristic information includes at least one of a front corner pose, a rear corner pose, and a rear view mirror pose, and the method includes:

s1082, the vehicle logo information of the obstacle vehicle is not acquired, and the front corner pose, the rear corner pose and/or the rearview mirror pose of the obstacle vehicle are acquired based on the SLAM module of the vehicle.

s1083, when the number of the grounding points of the wheels of the obstacle vehicle is smaller than or equal to 1, acquiring the front corner pose, the rear corner pose and/or the rearview mirror pose of the obstacle vehicle based on the SLAM module of the vehicle.

As is apparent from the above description of the embodiments, it is preferable that the detection of the obstacle vehicle of the present disclosure is performed by means of the ground contact points of the wheels of the obstacle vehicle together with the emblem information of the obstacle vehicle and the like.

In some embodiments of the present disclosure, the present disclosure first looks up a corresponding vehicle model from a vehicle model library 1012 stored on a memory by means of a vehicle logo (exemplarily, a popular, great wall, great ann, biedi, etc. vehicle logo), a wheel spacing (obtained from a wheel contact point), and determines a specific location of a vehicle model set by means of a vehicle orientation and a wheel contact point location detected by a camera.

The method for mechanically applying and positioning the vehicle model for the obstacle vehicle with the common vehicle type has the advantages that the position of the rearview mirror of the obstacle vehicle can be accurately confirmed, and the obstacle avoidance judgment is avoided mistakenly due to the fact that the rearview mirror is difficult to detect and the like.

The narrow road passing processing method S100, S110 according to the preferred embodiment of the present disclosure, calculating the shortest distance between an obstacle vehicle and a narrow road based on the feature information of the obstacle vehicle to preliminarily determine whether the host vehicle can pass through without collision, includes:

the type of the narrow road is a straight narrow road, and the obstacle vehicle is located on a first side of the straight narrow road, then the minimum value of the distance between the front corner and the rear corner of the obstacle vehicle and the distance between the vehicle rearview mirror and the edge line (line segment) of a second side of the straight narrow road opposite to the first side is obtained as the shortest distance.

In the case of a straight narrow road, the shortest distance between the obstacle vehicle and the straight narrow road is generally the smallest of the distance between the front corner of the obstacle vehicle, the rear corner of the vehicle, or the edge line of the mirror on the other side (i.e., the second side) of the straight narrow road, due to the shape of the obstacle vehicle.

According to the above description of the present disclosure, the present disclosure can calculate the front corner, rear corner, and side of the protruding point of the vehicle mirror to the other side of the straight narrow road of the obstacle vehicle by means of the vehicle model or SLAMDistance between edge lines

Taking the minimum value of the three values as the shortest passable distance s.

According to the narrow road passing processing method S100 of the preferred embodiment of the present disclosure, when one or more of the intersection points of the perpendicular line made to the edge line (line segment) of the straight narrow road by the front corner, the rear corner, and the mirror to the edge line of the straight narrow road and the edge line do not fall on the edge line of the straight narrow road, the shortest distance is obtained based on the following steps:

identifying intersections that do not lie on edge lines

Taking the minimum of the distances(s) between the intersection points and the edge lines that are not identified ^R ) As a first minimum value;

acquiring straight narrow road edge line end points (straight narrow road edge line end points P, if the straight narrow road is short and two straight narrow road edge line end points can be observed at the periphery of the obstacle vehicle at the same time, acquiring two end points P ₁ 、P ₂ ) Calculating the shortest distance(s) between the end point (P) of the edge line of the straight narrow road and the obstacle vehicle ^P The distance direction is perpendicular to the central axis direction of the barrier vehicle, and the central axis direction is closer to the edge line direction of the straight narrow road as the central axis direction of the barrier vehicle because the barrier vehicle is provided with a front central axis, a rear central axis and a left central axis;

if the perpendicular segment corresponding to the shortest distance between the edge line end point (P) of the straight narrow road and the barrier vehicle can fall on the barrier vehicle, the shortest distance(s) is reserved ^P ) And as a second minimum value;

will be the first minimum value s ^R And a second minimum value s ^P And comparing, and taking the smaller value as the shortest distance between the obstacle vehicle and the narrow road.

In some embodiments of the present disclosure, since the narrow road is essentially a line segment of a certain length, not extending indefinitely, it is inevitable that, when calculating the distance from the protrusion point of an obstacle (the front corner, the rear corner, and the protrusion point of the rear view mirror) to the narrow road segment, the perpendicular from the protrusion point to the narrow road segment may not fall on the narrow road segment, as shown in fig. 3.

In some embodiments of the present disclosure, if a perpendicular line segment corresponding to a shortest distance between a straight narrow road edge line end point (P) and an obstacle vehicle fails to fall on the obstacle vehicle, the following processing is performed:

calculating the length of the connecting line between the end point (P) of the edge line of the straight narrow road and the front corner, the rear corner and the rear-view mirror (the protruded point) of the obstacle vehicle

Taking the minimum value as the second minimum value.

In some embodiments of the present disclosure, the second minimum value is recalculated if the perpendicular segment corresponding to the shortest distance between the straight narrow road edge line end point (P) and the obstacle vehicle fails to fall on the obstacle vehicle.

In some embodiments of the present disclosure, S110, calculating a shortest distance between the obstacle vehicle and the narrow road based on the feature information of the obstacle vehicle to preliminarily determine whether the host vehicle can pass through without collision, includes:

the type of the narrow road is a broken line narrow road, and the barrier vehicles are positioned in the break point area of the broken line narrow road, so that the broken line narrow road is used as two crossed straight line narrow roads R ₁ 、R ₂ Respectively judging whether the vehicle can pass through without collision.

Fig. 4 shows a case where there is an obstacle vehicle on one side of the narrow road of the broken line (in the broken point region) in the present embodiment.

In some embodiments of the present disclosure, if there is an obstacle vehicle in each part of the broken-line narrow road, the processing is performed based on the above-described method of acquiring the shortest distance between the obstacle vehicle and the narrow road when there is an obstacle vehicle in the straight narrow road.

In some embodiments of the disclosure, the type of the narrow road is a broken line narrow road, and the obstacle vehicle is located on the broken line narrow roadIn the region of the break point, the narrow road of the broken line is taken as two crossed straight narrow roads R ₁ 、R ₂ Whether this car can pass through without the collision is judged respectively, include:

Due to the vehicle geometry and the broken line narrow road geometry, in some embodiments of the present disclosure, the shortest distance between the obstacle vehicle and the narrow road may occur at the shortest distance from the narrow road break point to the obstacle vehicle in addition to the shortest distance from the front corner, rear view mirror of the obstacle vehicle to the narrow road.

In some embodiments of the present disclosure, the front corner, the rear corner and the protruding point of the rear view mirror of the obstacle vehicle are respectively calculated to each straight narrow section R by means of the vehicle model or SLAM ₁ 、R ₂ Is a distance of

And &>

Taking the shortest distance from the obstacle vehicle to each section of the straight line narrow section->

And &>

Then calculating the shortest distance s from the narrow road break point to the obstacle vehicle ^P (ii) a Taking or combining>

s ^P The minimum value of (1) is taken as the shortest distance between the obstacle vehicle and the narrow road.

In this disclosure, for all shortest distances s in a narrow road segment _n (n =1,2,3, \ 8230;) including a plurality of the same shortest distances between the same obstacle vehicle and the narrow road and the shortest distances between different obstacle vehicles and the narrow road, and the shortest distance closest to the host vehicle is taken as s ₁ And the numbering is automatically increased by taking 1 as a unit until the narrow-road passing module is exited.

s _n Including the length, direction, and coordinates of the two ends of the shortest distance line segment, in this disclosure, for the sake of brevity, it will be shown that s is used _n Means of generation s _n The case of a value.

In the present disclosure, if a perpendicular line from a protruding point of an obstacle vehicle to a narrow road segment cannot fall on the narrow road segment, or a perpendicular line from a broken point of the narrow road to the obstacle vehicle cannot fall on the obstacle vehicle, the method described above in the present disclosure for acquiring the shortest distance between the obstacle vehicle and the narrow road when the obstacle vehicle exists in the straight narrow road segment is used for processing.

In some embodiments of the present disclosure, it is noted that a situation that does not occur in a straight narrow road may occur in a broken-line narrow road, as shown in fig. 4, although the obstacle vehicle stops at an intersection point (i.e., a break point region) of the broken-line narrow road, the protrusion point of the obstacle vehicle and a certain narrow road R ₁ All the vertical lines of (A) cannot fall on R ₁ In this case, R ₁ The shortest distance to the obstacle vehicle is at R ₁ At the end point near the vehicle end of the obstacle, i.e. R ₁ And R ₂ At the intersection of (a) and (b), therefore, for

Is all set to->

Is equivalent to @, i.e., all are flagged as erroneous values>

Infinity, proceed only>

And s ^P To be compared.

Fig. 5 shows a case where there is an obstacle vehicle on one side in a curved narrow road.

In the present embodiment, it is preferable that the narrow road R in the circular safety threshold range including the obstacle vehicle is taken with the safety threshold of 5 meters as the radius with the obstacle vehicle as the center, and the narrow road R in the circular safety threshold range is evenly divided into the approximately straight road sections of which the length is 0.5m (adjustable) because the curved narrow road is smooth (the length of the narrow road R is not necessarily an integral multiple of 0.5, and therefore the length of the last short road section is (0,0.5)]In range), i.e., R = { R = { R =) ₁ ,R ₂ ,...,R _n Will each R therein _x (x is more than or equal to 1 and less than or equal to n) is taken as a section of straight narrow road, the method for acquiring the shortest distance between the obstacle vehicle and the narrow road when the obstacle vehicle exists in the straight narrow road section described in the disclosure is adopted for processing, and each section R is calculated _x Shortest distance to obstacle vehicle

Comparing n->

The minimum value of the length values is taken as a curve narrow road and an obstacleThe shortest distance s of the vehicle.

the obstacle vehicles are positioned on two sides of the narrow road, the shortest distance between each obstacle vehicle on the first side of the narrow road and each obstacle vehicle on the second side of the narrow road is obtained, and then the shortest distance between the obstacle vehicles on the two sides of the narrow road, namely the shortest passable distance(s), is obtained.

Fig. 6 shows a case where there are obstacle vehicles on both sides of a narrow road.

In some embodiments of the present disclosure, for the case where there are obstacle vehicles on both sides of the narrow road, the rear angle, the front angle, and the shortest distance s from the vehicle mirror to the obstacle vehicle on the second side (right side) of the narrow road are calculated ^C1 Calculating the rear angle and the front angle of the vehicle with the obstacle on the second side (right side) of the narrow road and the shortest distance from the rearview mirror to the vehicle with the obstacle on the first side (left side) of the narrow road

Pick the shortest distance->

To the shortest distance s ^C2 As two obstacle vehicles (C) ₁ 、C ₂ ) The shortest distance s (i.e., the shortest passable distance) therebetween.

If it is obtained

And/or>

Are all->

That is, the corresponding values are infinite, and as shown in fig. 6, the rear angles of the left and right obstacle vehicles are calculated respectivelyThe distance between the front corner of the vehicle and the connecting line between the rearview mirrors (9 in total), and taking the minimum value as two obstacle vehicles (C) ₁ 、C ₂ ) The shortest distance s therebetween.

Fig. 7 shows the case of a multi-obstacle vehicle.

If more than three obstacle vehicles in the narrow road are in a short distance, namely the obstacle vehicles on two sides of the narrow road stop in a non-one-to-one correspondence, as shown in fig. 7, the shortest distance(s) between the obstacle vehicles on two sides is calculated respectively ₁ 、s ₂ )。

Preferably, the two obstacle vehicles C closest to the vehicle are first paired ₁ 、C ₂ Making the shortest passable distance s ₁ To two obstacle vehicles C ₂ 、C ₃ Making the shortest passable distance s ₂ Is calculated and is given to s ₁ 、s ₂ The involved calculation content is reserved. Based on the above description "for all shortest distances s in a narrow road _n (n =1,2,3, \ 8230;) including a plurality of the same shortest distances between the same obstacle vehicle and the narrow road and the shortest distances between different obstacle vehicles and the narrow road, and the shortest distance closest to the host vehicle is taken as s ₁ And the number is increased by 1 as a unit until the narrow road passing module exits, and the processing is carried out in the same way.

According to a preferred embodiment of the present disclosure, the type of the narrow road of the present disclosure is judged based on the panoramic top view.

The narrow road passing processing method S100, S112 according to the preferred embodiment of the present disclosure preliminarily determines whether the vehicle can pass through without collision based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety factor, and includes:

if so, preliminarily judging that no collision can pass; if not, the collision-free passing cannot be judged.

In some embodiments of the present disclosure, the determining again whether the vehicle can pass through without collision based on at least the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information includes:

In the present disclosure, it is possible to assume that a is a width where the vehicle does not include an external mirror, and b is a width where the vehicle includes an external mirror.

In the present disclosure, the shortest distance(s) between the obstacle vehicle and the narrow road _n ). Theoretically, when s _n When the vehicle is higher than a, the vehicle can pass through a narrow road without collision, and when s exists _n When a is less than or equal to a, the vehicle can not pass through the narrow road without collision. But there is still a certain margin to ensure safety during actual automatic driving. In the present disclosure, a safety factor xi is set, and s is required _n The value of (b) is greater than the product of the safety factor and the width of the vehicle, so as to ensure safety. Therefore, when s _n If s is greater than xi · a, the narrow-path traffic module 1004 will continue to make an accurate determination _n ξ · a is less than or equal to, the narrow road passing module 1004 judges that the obstacle cannot be avoided, can generate a voice signal to inform the driver, and exits the automatic driving mode.

According to the narrow road traffic processing method S100 of the preferred embodiment of the present disclosure, determining whether or not a collision-free traffic can be passed based on the shortest passable distance between the obstacle vehicle and the narrow road, the own vehicle feature width information, and the narrow road type includes:

if so, judging that the vehicle can run along the straight narrow road and pass through the straight narrow road; if not, it is determined that the host vehicle stows the exterior mirror to be able to travel along the straight narrow road through the straight narrow road.

In the present disclosure, in the case of a straight narrow road, the shortest distance(s) between the obstacle vehicle and the narrow road _n ) When all the shortest distances s _n When the vehicle is larger than xi · b, the vehicle can run through a straight narrow road in the direction of the straight narrow road; when all s _n Greater than xi a, but xi a < s _n When the front view mirror is less than xi · b, the vehicle can drive in the direction of a straight narrow road and pass through the straight narrow road after retracting the outer rearview mirror; if s is present _n And xi · a is less than or equal to, the narrow road passing module 1004 judges that the obstacle can not be avoided, can generate a voice signal to inform a driver, and exits the automatic driving mode.

if the vehicle is positioned on the inner side of a curve (the side to which the centripetal force is directed when the vehicle turns is the inner side of the curve), a perpendicular line is drawn from the narrow road break point position along the direction (s direction) vertical to the axis of the obstacle vehicle, the width of the vehicle without an external rearview mirror/the width of the vehicle with the external rearview mirror (a or b) is taken as a position p on the perpendicular line, and line segments with the length of 1/2 of the vehicle length are respectively drawn from the position p to the front and the rear of the perpendicular line;

and judging whether the line segments are intersected with the narrow road or not, and judging that the vehicle can run through the narrow road (entering an obstacle avoidance route planning step).

Because the vehicle has a certain length and can not be bent or bent, the situation that the vehicle with obstacles on one side or the vehicles with a plurality of obstacles on two sides of the nonlinear narrow road in the narrow road passing is not only ensured _n The vehicle can pass through more than xi.a, mainly because the distance between a plurality of shortest distances is short and when the vehicle needs to turn, the vehicle cannot be ensured to be vertical to s _n-1 And s _n And (4) passing. Therefore, additional restraint is required for the case that the obstacle vehicles stop at the break point in the broken line narrow road, the curve narrow road has more than two obstacle vehicles, both sides of the curve narrow road have obstacle vehicles, and the number of the obstacle vehicles is more than two.

In a broken line narrow road, the obstacle vehicle is stoppedWhen the vehicle is not at the folding point, the vehicle can be regarded as an obstacle and stopped on the straight narrow road, and the vehicle is processed in the straight narrow road in the embodiment; however, when the obstacle vehicle is located right at the position of the break point, referring to fig. 8, since the road turning is large, there is a certain chance even if s is secured _n If the vehicle is greater than ξ · a, there is still a certain probability that the vehicle cannot pass through without collision due to a collision road edge or the like, and therefore further conditions need to be satisfied in order to ensure that the vehicle passes through a narrow road without collision.

Referring to fig. 8, if the vehicle is located inside a curve (the side to which the centripetal force is directed when the vehicle turns is the inside of the curve) rather than the obstacle vehicle, as shown in fig. 8, the length a or b is taken from the end of the narrow road along the direction s as the position p, a segment with the length of 1/2 of the length of the vehicle is respectively taken from the point p to the front and the back perpendicular to the straight line s, the coordinates of the end point of the segment and the coordinates of the narrow road are compared, and if the segment does not intersect with the narrow road, the obstacle avoidance route is planned; otherwise, the vehicle is judged not to pass through smoothly, and a voice signal can be generated to inform the driver and quit the automatic driving.

According to the narrow road traffic processing method S100 of the preferred embodiment of the present disclosure, determining whether or not a collision-free traffic can be passed based on the shortest passable distance between the obstacle vehicle and the narrow road, the own vehicle characteristic width information, and the narrow road type includes:

the narrow road type is a broken line narrow road, the obstacle vehicle is located at the break point position of the broken line narrow road, and if the vehicle is located on the outer side of the curve compared with the obstacle vehicle, the judgment is carried out based on the following judgment logic:

When the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning left, the line segment s ₁ Right end point is taken as p ₁ Dot, by line segment s ₂ Left end endpoint as p ₂ Point; when the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning to the right, with line segment s ₁ Left end endpoint as p ₁ Dot, with lineSegment s ₂ The right end point is taken as p ₂ Point;

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

The method comprises the steps of obtaining the length L of a diagonal line of a rectangular frame which takes the length of a vehicle as a long side and does not contain the width of a rearview mirror as a short side ₁ (ii) a The length L of the diagonal line of a rectangular frame with the length of the vehicle as the long side and the width of the vehicle containing a rearview mirror as the short side is obtained ₂ ；

Comparison S _L And L ₁ 、L ₂ If s is the value of _n Is > ξ · b and s _L ＞ξ·L ₂ Judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi. B or xi. L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi a or s is less than or equal to _L ≤ξ·L ₁ Then the vehicle cannot pass through without collision.

If the host vehicle is located outside of the curve than the obstacle vehicle, s has been obtained above for this disclosure ₁ 、s ₂ (the broken line narrow roads are regarded as two straight line narrow roads, so that the same obstacle vehicle will have two shortest distances corresponding to "different" narrow roads, which coincide to a shortest distance unless the shortest distance is located at the intersection of two straight lines):

as shown in FIG. 9, if the host vehicle moves from the position s ₁ To s ₂ If a left turn is required, s is obtained ₁ The end point of the right end of the line segment is taken as p ₁ ，s ₂ The end point at the left end of the line segment is taken as p ₂ Obtaining p ₁ 、p ₂ Length S of the connecting wire _L 。

If the host vehicle is driven by the vehicle ₁ To s ₂ If a turn to the right is required, s is obtained ₁ The end point at the left end of the line segment is taken as p ₁ ，s ₂ The end point of the right end of the line segment is taken as p ₂ . Obtaining p ₁ 、p ₂ Length S of the connecting wire _L 。

Obtaining a rectangular frame with the length of the vehicle as the long side and the width a of the vehicle without a rearview mirror as the short sideDiagonal length L of ₁ (ii) a The length L of the diagonal line of a rectangular frame taking the length of the vehicle as the long side and the width b of the rearview mirror as the short side ₂ 。

Comparison S _L And L ₁ 、L ₂ The magnitude of the value of (c).

If s _n B and s > xi _L ＞ξ·L ₂ Judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi. B or xi. L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi a or s is less than or equal to _L ≤ξ·L ₁ Then, the vehicle cannot pass through without collision, and a voice signal can be generated to notify the driver and quit the automatic driving.

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

The moment taking the length of the vehicle as the long side and the width of the vehicle without the rearview mirror as the short side is obtainedDiagonal length L of the frame ₁ (ii) a The length L of the diagonal line of a rectangular frame with the length of the vehicle as the long side and the width of the vehicle containing a rearview mirror as the short side is obtained ₂ ；

Comparison S _L And L ₁ 、L ₂ If s is the value of _n B and s > xi _L ＞ξ·L ₂ Judging that the vehicle can pass through the narrow road without obstacles; for all s _n Is > ξ. A and s _L ＞ξ·L ₁ If xi a < s exists _n < xi · b or xi · L ₁ ＜s _L ＜ξ·L ₂ Then the vehicle can pass through a narrow road by retracting the outer rearview mirror; if s is present _n Xi a or s is less than or equal to _L ≤ξ·L ₁ Then the vehicle cannot pass through without collision.

for various narrow road types, if obstacle vehicles are arranged on both sides of the narrow road and more than two obstacle vehicles are arranged on one side of the narrow road, the judgment is carried out based on the following judgment logics:

obtaining the shortest distance line segment s between the first barrier vehicle and the second barrier vehicle ₁ And acquiring the shortest distance line segment s between the third barrier vehicle and the second barrier vehicle ₂ The first barrier vehicle and the third barrier vehicle are on the same side of the narrow road, and the second barrier vehicle is on the other side of the narrow road;

when the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning left, the line segment s ₁ The right end point is taken as p ₁ Point, by line segment s ₂ Left end endpoint as p ₂ Point; when the vehicle needs to pass through the line segment s ₁ 、s ₂ When turning to the right, with line segment s ₁ Left end endpoint as p ₁ Point, by line segment s ₂ Right end point is taken as p ₂ Point;

obtaining p ₁ Point and p ₂ Length S of dot chain line _L ；

In a curved narrow road or a narrow road with obstacle vehicles on both sides, due to the existence of multiple obstacle vehicles, when the multiple obstacle vehicles are close to the shortest distance between the narrow roads, turning is needed possibly to cause that the vehicle cannot be perpendicular to s _n-1 And s _n The situation of passing. Thus in addition to the basic s _n ξ · b, and a comparison s _L And L ₁ 、L ₂ The magnitude of the value of (c). s is _L As in the above embodiments _L The same method is used.

As shown in fig. 10 and 11, if the host vehicle is driven by the vehicle s ₁ To s ₂ If a left turn is required, s is obtained ₁ The end point of the right end of the line segment is taken as p ₁ ，s ₂ The end point at the left end of the line segment is taken as p ₂ Obtaining p ₁ 、p ₂ Length S of the connecting wire _L 。

If the host vehicle is composed of ₁ To s ₂ If a turn to the right is required, s is obtained ₁ The end point at the left end of the line segment is taken as p ₁ ，s ₂ The end point of the right end of the line segment is taken as p ₂ . Obtaining p ₁ 、p ₂ Length S of the connecting wire _L 。

Obtaining the pair of rectangular frames which take the length of the vehicle as the long side and the width a of the rearview mirror as the short sideAngular length L ₁ (ii) a The length L of the diagonal line of a rectangular frame taking the length of the vehicle as the long side and the width b of the rearview mirror as the short side ₂ 。

Comparison S _L And L ₁ 、L ₂ The magnitude of the value of (c).

In some embodiments of the present disclosure, when there is an obstacle vehicle that cannot detect a corresponding model, the present disclosure uses the salient feature points of the obstacle vehicle obtained by the SLAM to connect the salient feature points with the front corner, the rear corner, and the rearview mirror of other obstacle vehicle models, and takes the shortest value of the connection distance as the shortest passable distance s. When a plurality of obstacle vehicles cannot detect the corresponding models, the salient feature points of the obstacle vehicles acquired by the SLAM are used for connecting the salient feature points with the salient feature points of other obstacle vehicles in a pairwise correspondence manner, and the value with the shortest connecting line distance is taken as the shortest passable distance s.

According to the narrow road traffic processing method S100, S114 of the preferred embodiment of the present disclosure, when it is determined again that the host vehicle can pass through without collision, the obstacle avoidance route planning is performed, and includes:

and planning the obstacle avoidance route based on the shortest passable distance.

In the disclosure, for a narrow broken line road, when an obstacle vehicle is positioned at a broken point and a vehicle is positioned on the inner side of a curve, the obstacle avoidance path direction is planned along the direction of the shortest distance vertical line, and the vehicle drives close to the inner side of the curve; when the obstacle avoidance line passes through the narrow road of the broken line except the condition, and all the straight narrow roads and the curved narrow roads, the obstacle avoidance line direction is ruled along the narrow road directionMarking the shortest passable distance s between the center of the rear axle of the vehicle and the shortest passable distance _n Are coincident.

In the disclosure, when there are obstacle vehicles on both sides of a narrow road, if the distance between the obstacle vehicles on both sides becomes narrower along with the advancing direction (one end point of the shortest distance is located at the front corner or the rear view mirror of a certain obstacle vehicle), the direction of the obstacle avoiding route is planned along the parking direction of the obstacle vehicle farther from the host vehicle (see fig. 12, where "obstacle vehicle farther from the host vehicle" refers to the vehicle farther from the host vehicle, such as c, of the two obstacle vehicles ₁ And c ₂ C in (1) ₂ ，c ₂ And c ₃ C in (1) ₃ ) (ii) a If the distance between the two obstacle vehicles is widened along with the advancing direction, the obstacle avoiding route is planned along the stopping direction of the obstacle vehicles close to the vehicle. S passing through the tail of the vehicle at right angle _n Then, based on the general vehicle control algorithm in the prior art, the direction of the vehicle is adjusted, and the vehicle arrives at the head of the vehicle by s _n+1 Front, the direction of the vehicle is made perpendicular to s _n+1 。

In the disclosure, for the obstacle vehicles using the SLAM to detect the feature points, the obstacle avoidance route is planned to be always perpendicular to the nearest shortest passable distance s _n The direction of the connection; and pass through the current s at the tail of the vehicle _n When the vehicle head reaches the next s _n+1 Previously, a uniform and smooth transition was made to the next s based on the general vehicle control algorithms in the prior art _n+1 Is transferred in the vertical direction.

And when the vehicle finishes the current narrow road passing along the obstacle avoidance planning and the rough measurement module finds that no obstacle vehicle exists in the measurable range, judging that the obstacle avoidance is finished and closing the narrow road passing module. If a new obstacle is found in the obstacle avoidance passing process of the narrow passage passing module which is not closed, the calculation of the new obstacle avoidance path is still carried out based on the determined world coordinate system when the narrow passage passing module is opened.

As can be seen from the above description, the narrow passage processing method of the present disclosure has the following technical advantages/effects: the rough detection narrow road is combined with the detailed narrow road passing module, so that the calculation amount in the automatic driving process of the vehicle can be reduced; the vehicle position and pose can be accurately determined by means of the IMU and the wheel speed, and the vehicle positioning can be completed under the condition that satellite positioning signals are missing or weak, so that the vehicle positioning precision is improved; the system comprises a plurality of frames of overlapping looking-around fish-eye cameras, and can overlap and splice a plurality of target images shot around the vehicle within a period of time so as to improve the accuracy and the applicability of the detection of the obstacle vehicle; the vehicle-mounted all-round system is preferably used, images are obtained by using fisheye cameras arranged on the periphery of a vehicle body, all-round images are obtained through image distortion correction, inverse perspective transformation and image splicing fusion, the visual field of the all-round system is concentrated in a near range of the periphery of the vehicle, clear lane line information on two sides and in the front and back direction of the vehicle can be provided, and linear characteristics are presented in the visual field range, so that type identification is facilitated (most of the existing vision-based lane line detection adopts a camera with a forward visual angle, although the visual range is far, the resolution of a far lane line is low, the far lane line is easily influenced by backlight and shielded by the vehicle, and the lane lines on two sides of the vehicle are not easily interfered by other vehicles relative to the front of the vehicle); the obstacle vehicles are classified by means of the vehicle model library, so that accurate data of the vehicles can be obtained in a more rapid, effective and low-cost mode, and the problem of scratch caused by low detection precision of the vehicle rearview mirrors is effectively solved; by means of a vehicle model, a SLAM feature point, a narrow road model and the like, the shortest passing distance between obstacles, the position and the passing direction of the obstacles are accurately acquired through a mode of classifying vehicles based on the narrow road and the obstacles in multiple situations.

In addition, in order to solve the scratch problem caused by the problem that the vehicle cannot move straight when the vehicle width is close to the narrow road width, the vehicle width and narrow road width are combined to carry out classification processing on different narrow road conditions, and the vehicle length, width, narrow road position, pairwise shortest passing distance and other data are combined with a mode of allowing the vehicle to retract the rearview mirror to avoid obstacles, so that the vehicle can be accurately judged to pass through the narrow road without collision, and the vehicle driving decision is more reliable.

Fig. 15 is a block diagram illustrating the structure of the narrow-way traffic module 1004 according to an embodiment of the present disclosure.

The narrow passage processing apparatus 1000 according to an embodiment of the present disclosure, with reference to fig. 14, includes:

the rough measurement module 1002 is used for judging whether narrow road and/or narrow road obstacles (vehicles and the like) exist in front of the vehicle or not based on preset narrow road characteristics, and if so, the narrow road passing module 1004 of the vehicle is started;

a narrow pass module 1004, the narrow pass module 1004 being activated in response to the rough test module 1002;

a pose acquisition module 1006, the pose acquisition module 1006 acquiring the start time (t) of the vehicle at the narrow passage module 1004 _on ) The pose of (a);

a panoramic top view acquisition module 1008, wherein the panoramic top view acquisition module 1008 acquires frame images of the surroundings of the vehicle, which are acquired by each camera of the vehicle within a preset time length, by using the starting time of the narrow passage module 1004 as a reference time, so as to obtain a panoramic top view of the vehicle;

the obstacle detection module 1010 is used for detecting an obstacle vehicle based on the panoramic top view to acquire characteristic information of the obstacle vehicle, wherein the characteristic information comprises at least one of a front corner pose, a rear corner pose and a rearview mirror pose;

preferably, with reference to fig. 15, the narrow-road passing module 1004 comprises:

the first judging module 4002, the first judging module 4002 calculates the shortest passable distance between the obstacle vehicle and the narrow road (i.e. the narrowest length that the vehicle needs to pass through in the process of passing through the narrow road) based on the feature information of the obstacle vehicle, so as to preliminarily judge whether the vehicle can pass through without collision;

a second determination module 4004, where the second determination module 4004, when the first determination module 4002 preliminarily determines that the vehicle can pass through without collision, determines whether the vehicle can pass through without collision again based on at least a shortest passable distance between the obstacle vehicle and a narrow road and vehicle characteristic width information, where the vehicle characteristic width information includes a width that the vehicle includes an external rearview mirror and/or a width that the vehicle does not include an external rearview mirror;

when the second determination module 4004 determines that the vehicle can pass through without collision again, the route planning module 4006 performs obstacle avoidance route planning.

Preferably, calculating the shortest passable distance between the obstacle vehicle and the narrow road based on the characteristic information of the obstacle vehicle to preliminarily judge whether the own vehicle can pass through without collision includes:

In some embodiments of the present disclosure, the obstacle detection module 1010 is included in the narrow-way passage module 1004.

In some embodiments of the present disclosure, the panoramic top view acquisition module 1008 and the obstacle detection module 1010 are included in the narrow road passage module 1004.

The narrow-passage processing device 1000 may include corresponding modules for performing each or several steps in the above-described flowcharts. Thus, each step or several steps in the above-described flow charts may be performed by a respective module, and the apparatus may comprise one or more of these modules. The modules may be one or more hardware modules specifically configured to perform the respective steps, or implemented by a processor configured to perform the respective steps, or stored within a computer-readable medium for implementation by a processor, or by some combination.

The hardware architecture may be implemented using a bus architecture. The bus architecture may include any number of interconnecting buses and bridges depending on the specific application of the hardware and the overall design constraints. The bus 1100 couples various circuits including the one or more processors 1200, the memory 1300, and/or the hardware modules together. The bus 1100 may also connect various other circuits 1400, such as peripherals, voltage regulators, power management circuits, external antennas, and the like.

The bus 1100 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one connection line is shown, but this does not indicate only one bus or one type of bus.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present disclosure includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the implementations of the present disclosure. The processor performs the various methods and processes described above. For example, method embodiments in the present disclosure may be implemented as a software program tangibly embodied in a machine-readable medium, such as a memory. In some embodiments, some or all of the software program may be loaded and/or installed via memory and/or a communication interface. When the software program is loaded into memory and executed by a processor, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the processor may be configured to perform one of the methods described above in any other suitable manner (e.g., by means of firmware).

The logic and/or steps represented in the flowcharts or otherwise described herein may be embodied in any readable storage medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.

For the purposes of this description, a "readable storage medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the readable storage medium include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM). In addition, the readable storage medium may even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in the memory.

It should be understood that portions of the present disclosure may be implemented in hardware, software, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

It will be understood by those skilled in the art that all or part of the steps of the method implementing the above embodiments may be implemented by hardware that is related to instructions of a program, and the program may be stored in a readable storage medium, and when executed, the program may include one or a combination of the steps of the method embodiments.

In addition, each functional unit in the embodiments of the present disclosure may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a readable storage medium. The storage medium may be a read-only memory, a magnetic or optical disk, or the like.

In the description of the present specification, reference to the description of "one embodiment/implementation", "some embodiments/implementations", "examples", "specific examples", or "some examples", etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment/implementation or example is included in at least one embodiment/implementation or example of the present application. In this specification, the schematic representations of the terms described above are not necessarily the same embodiment/mode or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments/modes or examples. Furthermore, the various embodiments/aspects or examples and features of the various embodiments/aspects or examples described in this specification can be combined and combined by one skilled in the art without conflicting therewith.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

It will be understood by those skilled in the art that the foregoing embodiments are merely for clarity of illustration of the disclosure and are not intended to limit the scope of the disclosure. Other variations or modifications may occur to those skilled in the art, based on the foregoing disclosure, and are still within the scope of the present disclosure.

Claims

1. A narrow-road traffic processing method is characterized by comprising the following steps:

acquiring the pose of the vehicle at the starting time of the narrow road passing module;

the method comprises the steps that starting time of a narrow-path passing module is used as reference time, frame images of the surrounding environment of a vehicle, collected by cameras of the vehicle in a preset time length, are obtained, and a panoramic top view of the vehicle is obtained;

if the vehicle is preliminarily judged to pass through without collision, judging whether the vehicle can pass through without collision again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information, wherein the vehicle characteristic width information comprises the width of the vehicle containing an external rearview mirror and/or the width of the vehicle not containing the external rearview mirror; and

when judging that the vehicle can pass through without collision again, planning an obstacle avoidance route;

the method for acquiring the panoramic top view of the vehicle by taking the starting time of the narrow-path passing module as the reference time and acquiring the frame images of the surrounding environment of the vehicle, which are acquired by each camera of the vehicle in a preset time length, comprises the following steps:

carrying out distortion removal on the frame image collected by each camera to obtain the frame image of each camera after distortion removal;

for each camera, the moment corresponding to the first frame image acquired by the camera is used as the initial moment, and the other frame images are subjected to image characteristic point position adjustment based on the IMU data and the wheel speed data of the camera;

spatially aligning all frame images of each camera after the time alignment to a world coordinate system by a camera coordinate system; and

synthesizing the frame images of each camera after the time alignment and the space alignment, and obtaining the panoramic top view by combining the pose of the vehicle at the starting time of the narrow-path traffic module;

wherein, calculate the shortest distance that can pass between barrier vehicle and the narrow road based on the characteristic information of barrier vehicle to whether preliminary judgement this car can pass through without the collision, include:

the type of the narrow road is a straight narrow road, and the obstacle vehicle is positioned on a first side of the straight narrow road, so that the minimum value of the front angle and the rear angle of the obstacle vehicle and the distance between the rearview mirror and the edge line of a second side, opposite to the first side, of the straight narrow road is obtained and used as the shortest passable distance;

when one or more of the intersection points of the perpendicular lines made by the front corner, the rear corner and the edge lines of the rearview mirror to the straight narrow road and the edge lines do not fall on the edge lines of the straight narrow road, acquiring the shortest passable distance based on the following steps:

marking the intersection points which do not fall on the edge lines;

and comparing the first minimum value with the second minimum value, and taking the smaller value as the shortest passable distance between the obstacle vehicle and the narrow road.

2. The narrow road passage processing method according to claim 1, wherein calculating the shortest passable distance between the obstacle vehicle and the narrow road based on the feature information of the obstacle vehicle to preliminarily determine whether the host vehicle can pass through without collision includes:

whether the vehicle can pass through without collision is preliminarily judged based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety factor.

3. The narrow-road passing processing method of claim 1 or 2, wherein whether a narrow road and/or a narrow-road obstacle exists in front of the vehicle is judged based on preset narrow-road characteristics, and if so, a narrow-road passing module of the vehicle is started, and the method comprises the following steps:

judging the distance between the vehicle and the narrow road and/or the narrow road obstacle when the narrow road and/or the narrow road obstacle exists; and

and when the distance between the vehicle and the narrow road and/or the narrow road obstacle is smaller than or equal to a preset threshold value, starting the narrow road passing module.

4. The narrow-road passage processing method according to claim 1, wherein acquiring the pose of the host vehicle at the start time of the narrow-road passage module includes:

acquiring the translation amount of the vehicle along the direction right ahead of the vehicle, the translation amount along the direction right left of the vehicle and the translation amount along the direction right above the vehicle, namely attitude change amount, in the time period from the acquisition time of the nearest satellite positioning data to the starting time of the narrow road passing module; and

5. The narrow-road passage processing method according to claim 1, wherein the obstacle vehicle detection is performed based on the panoramic top view, and feature information of the obstacle vehicle is acquired, the feature information including at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, and the method includes:

and calling a corresponding vehicle model, placing the vehicle model on the position of the obstacle vehicle in the panoramic top view based on the grounding point of the wheels of the obstacle vehicle and the direction of the head of the obstacle vehicle, and acquiring the front corner position, the rear corner position and/or the position of a rearview mirror of the obstacle vehicle based on the vehicle model.

6. The narrow-road passage processing method according to claim 1, wherein the obstacle vehicle detection is performed based on the panoramic top view, and feature information of the obstacle vehicle is acquired, the feature information including at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, and the method includes:

7. The narrow road traffic processing method according to claim 1, wherein the obstacle vehicle detection is performed based on the panoramic top view, and feature information of the obstacle vehicle is acquired, the feature information including at least one of a front corner pose, a rear corner pose, and a rear view mirror pose, and the method includes:

and acquiring the position of the front corner, the position of the rear corner and/or the position of a rearview mirror of the obstacle vehicle based on the SLAM module of the vehicle without acquiring the vehicle logo information of the obstacle vehicle.

8. The narrow-road passage processing method according to claim 1, wherein the obstacle vehicle detection is performed based on the panoramic top view, and feature information of the obstacle vehicle is acquired, the feature information including at least one of a front corner pose, a rear corner pose, and a rearview mirror pose, and the method includes:

9. The narrow road passage processing method according to claim 1, wherein if the perpendicular line segment corresponding to the shortest distance between the end point of the edge line of the straight narrow road and the obstacle vehicle fails to fall on the obstacle vehicle, the following processing is performed:

10. The narrow-road passage processing method according to claim 9, wherein the type of the narrow road is determined based on the panoramic top view.

11. The narrow road passing processing method according to claim 2, wherein preliminarily judging whether the vehicle can pass through without collision based on the shortest passable distance between the obstacle vehicle and the narrow road and a preset safety factor comprises:

obtaining the value of the product of the width of the vehicle and a preset safety factor;

judging whether the shortest passable distance between the barrier vehicle and the narrow road is greater than or equal to the value of the product of the width of the vehicle without the outside rearview mirror and a preset safety factor or not;

12. The narrow road passage processing method according to claim 2 or 11, wherein the re-determination of whether or not the collision-free passage is possible based on at least the shortest passable distance between the obstacle vehicle and the narrow road and the own vehicle feature width information includes:

13. A narrow-passage processing device, comprising:

a narrow passage module activated in response to the rough measurement module;

the narrow passage module comprises:

the second judgment module is used for judging whether the vehicle can pass through without collision again at least based on the shortest passable distance between the obstacle vehicle and the narrow road and the vehicle characteristic width information when the first judgment module preliminarily judges that the vehicle can pass through without collision, wherein the vehicle characteristic width information comprises the width of the vehicle containing an external rearview mirror and/or the width of the vehicle not containing the external rearview mirror;

the route planning module is used for planning an obstacle avoidance route when the second judging module judges that the vehicle can pass through without collision again;

the panoramic top view acquiring module acquires frame images of the surrounding environment of the vehicle, which are acquired by cameras of the vehicle within a preset time length, by taking the starting time of the narrow-path passing module as a reference time so as to acquire the panoramic top view of the vehicle, and the panoramic top view acquiring module comprises:

marking the intersection points which do not fall on the edge lines;

14. The narrow-road passage processing device according to claim 13, wherein calculating a shortest passable distance between the obstacle vehicle and the narrow road based on the feature information of the obstacle vehicle to preliminarily determine whether the host vehicle can pass through without collision includes:

15. The narrow-passage processing apparatus according to claim 13, wherein the obstacle detection module is included in the narrow-passage module.

16. The narrow-passage processing device of claim 13, wherein the panoramic top view acquisition module and the obstacle detection module are included in the narrow-passage module.

17. An electronic device, comprising:

a memory storing execution instructions; and

a processor executing execution instructions stored by the memory, causing the processor to perform the method of narrow-road passage processing of any one of claims 1 to 12.

18. A readable storage medium, wherein the readable storage medium stores therein execution instructions, and the execution instructions are executed by a processor to implement the narrow-road traffic processing method according to any one of claims 1 to 12.

19. An automobile, comprising: the device comprises a vehicle-mounted camera device, a display screen, a wheel speed meter, an inertia measuring device, a processor and a readable storage medium; the readable storage medium is characterized by storing an execution instruction, and the execution instruction is used for realizing the narrow-path traffic processing method in any one of claims 1 to 12 when being executed by a processor.

20. An automobile, comprising: the device comprises a vehicle-mounted camera device, a display screen, a wheel speed meter, an inertia measuring device, a processor and a readable storage medium; the narrow-path traffic processing device according to any one of claims 13 to 16, wherein the narrow-path traffic processing device is stored in the readable storage medium by executing instruction modules with a computer program.