CN115092185A - Dynamic obstacle avoidance method and device for automatic driving vehicle, vehicle and storage medium - Google Patents

Abstract

The application relates to the technical field of automobiles, in particular to a dynamic obstacle avoidance method and device for an automatic driving vehicle, the vehicle and a storage medium, wherein the method comprises the following steps: acquiring cloud data of a current driving road section of an automatic driving vehicle; identifying the actual road condition of the current driving road section according to the cloud data, and judging whether an abnormal road surface with the road surface flatness smaller than a preset value exists in the current driving road section according to the actual road condition; and when the current running road section has an abnormal road surface, controlling the vehicle to execute a corresponding obstacle avoidance action when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value. Therefore, the problems that an automatic driving vehicle in the related art cannot avoid a front abnormal road surface in advance, driving safety is reduced, user experience is poor and the like are solved.

Description

Dynamic obstacle avoidance method and device for automatic driving vehicle, vehicle and storage medium

Technical Field

The present application relates to the field of automotive technologies, and in particular, to a dynamic obstacle avoidance method and apparatus for an autonomous vehicle, a vehicle, and a storage medium.

Background

Along with the improvement of the intelligent degree of the automobile, the configuration of a front sensor and an actuator of the automobile is more and more abundant, the self-automobile perception capability is more and more strong, and along with the improvement of the calculation power of the intelligent automobile, the environment information which can be processed by the automobile is more abundant, and the more subtle road environment can be identified. Meanwhile, with the development of intelligent networking technology and a big data platform, the automobile has richer communication capacity, can realize automobile-automobile communication and automobile-cloud communication, and has stronger and stronger interaction capacity.

The automatic driving vehicle based on the single intelligent vehicle can realize the detection of the road environment, can identify the surrounding target vehicles, pedestrians, road information and the like, but is limited by the detection range of the sensor, still has the sensing capability limit for the road environment at a farther distance, particularly has the sensing capability limit for the road environment change caused by road construction and road structure damage, the vehicle cannot sense in time, can be identified only when the vehicle approaches or passes, and brings great limitation to the intelligent decision making of the automatic driving vehicle, because the timely road information cannot be sensed, the vehicle cannot make more intelligent decision at the first time, if the vehicle runs through the road construction or the road potholes, the intelligent vehicle can only find the road potholes when the vehicle runs through the road potholes, at the moment, the vehicle can bring great hidden danger to the driving stability of the vehicle due to the high-speed running through the road potholes, and even the vehicle tire burst can be caused, and serious traffic accidents can be caused when the accident is serious.

Disclosure of Invention

The application provides a dynamic obstacle avoidance method and device for an automatic driving vehicle, the vehicle and a storage medium, and aims to solve the problems that the automatic driving vehicle cannot avoid a front abnormal road surface in advance, driving safety is reduced, user experience is poor and the like in the related art.

The embodiment of the first aspect of the application provides a dynamic obstacle avoidance method for an automatic driving vehicle, which comprises the following steps: acquiring cloud data of a current driving road section of an automatic driving vehicle; identifying the actual road condition of the current driving road section according to the cloud data, and judging whether an abnormal road surface with the road surface evenness smaller than a preset value exists in the current driving road section according to the actual road condition; and when the abnormal road surface exists in the current running road section, controlling the vehicle to execute corresponding obstacle avoidance action when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value.

According to the technical means, the embodiment of the application can acquire the front lane condition based on the cloud data, and can avoid obstacles in advance when abnormal road surfaces exist in the front, so that the road condition in front of a vehicle can be acquired in advance based on the cloud data, the distance limit of vehicle perception is compensated, corresponding driving decisions are made in advance, driving safety is improved, automatic driving decisions are more intelligent, and use experience of users is improved.

Optionally, the controlling the host vehicle to execute a corresponding obstacle avoidance action when the actual distance between the host vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold includes: identifying whether a variable lane exists in the current driving road section; and if the lane changing exists, controlling the host vehicle to perform a lane changing action, otherwise, controlling the host vehicle to perform a braking and decelerating action.

According to the technical means, lane changing or braking deceleration can be selected according to the actual situation of the driving road section, so that the safety influence of abnormal road surfaces on driving is reduced through an optimal strategy, the intelligence of automatic driving decision is improved, and the use experience of a user is improved.

Further, after controlling the host vehicle to perform a lane change action, the method includes: acquiring the actual position of the abnormal road surface; acquiring an actual image of the road surface corresponding to the actual position; and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

According to the technical means, when the lane changing avoids the abnormal road surface, the road surface condition can be collected through the sensor of the vehicle, and when the road surface condition is not abnormal, the road surface information is uploaded to the cloud in time, so that other vehicles can realize driving decision according to the latest road surface condition, and invalid automatic driving decision is prevented from being carried out when other automatic driving vehicles pass through the road section again.

Optionally, after controlling the host vehicle to perform a braking deceleration action, the method includes: acquiring the actual position of the abnormal road surface; detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position; and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

According to the technical means, when the brake detection passes through the abnormal road surface, the road surface condition can be detected through the sensors of the vehicles, and when the road surface condition is detected to be abnormal, the road surface information is uploaded to the cloud end in time, so that other vehicles can realize driving decision according to the latest road surface condition, and invalid automatic driving decision is prevented from being carried out when other automatic driving vehicles pass through the road section again.

An embodiment of a second aspect of the present application provides a dynamic obstacle avoidance device for an autonomous vehicle, including: the acquisition module is used for acquiring cloud data of a current driving road section of the automatic driving vehicle; the identification module is used for identifying the actual road condition of the current driving road section according to the cloud data and judging whether an abnormal road surface with the road surface evenness smaller than a preset value exists in the current driving road section according to the actual road condition; and the control module is used for controlling the vehicle to execute corresponding obstacle avoidance actions when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value when the abnormal road surface exists in the current running road section.

Optionally, the control module is configured to: identifying whether a variable lane exists in the current driving road section; and if the lane changing exists, controlling the host vehicle to perform a lane changing action, otherwise, controlling the host vehicle to perform a braking and decelerating action.

Optionally, the method further comprises: a first detection module for acquiring an actual position of the abnormal road surface after controlling the host vehicle to perform a lane change action; acquiring an actual image of the road surface corresponding to the actual position; and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

Optionally, the method further comprises: the second detection module is used for acquiring the actual position of the abnormal road surface after controlling the host vehicle to execute lane changing action; detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position; and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end so as to update the state of the road surface corresponding to the actual position.

An embodiment of a third aspect of the present application provides a vehicle, comprising: the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the dynamic obstacle avoidance method of the automatic driving vehicle.

A fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor, so as to implement the dynamic obstacle avoidance method for an autonomous vehicle as described in the foregoing embodiments.

Therefore, the application has at least the following beneficial effects:

(1) according to the embodiment of the application, the lane condition in front can be obtained based on the cloud data, and the obstacle avoidance can be carried out in advance when an abnormal road surface exists in the front, so that the road condition in front of a vehicle can be obtained in advance based on the cloud data, the distance limit sensed by the vehicle is compensated, a corresponding driving decision is made in advance, the driving safety is improved, meanwhile, the automatic driving decision is more intelligent, and the use experience of a user is improved;

(2) according to the embodiment of the application, lane changing or braking deceleration can be selected according to the actual situation of the running road section, so that the safety influence of abnormal road surfaces on driving is reduced through an optimal strategy, the intelligence of automatic driving decision is improved, and the use experience of a user is improved;

(3) according to the embodiment of the application, when the lane change is carried out and an abnormal road surface is avoided, the road surface condition can be collected through the sensor of the vehicle, and when the road surface condition is not abnormal, the road surface information is uploaded to the cloud in time, so that other vehicles can conveniently realize driving decisions according to the latest road surface condition, and invalid automatic driving decisions are prevented from being carried out when other automatic driving vehicles pass through the road section again;

(4) the embodiment of the application can detect the road surface condition through the sensor of the vehicle when the braking detection passes through the abnormal road surface, and when the road surface condition is detected to be abnormal, the road surface information is uploaded to the cloud in time, so that other vehicles realize driving decision according to the latest road surface condition, and invalid automatic driving decision is carried out when other automatic driving vehicles pass through the road section again.

Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.

Drawings

The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

fig. 1 is a flowchart of a dynamic obstacle avoidance method for an autonomous vehicle according to an embodiment of the present application;

FIG. 2 is a diagram of a vehicle steering system architecture provided in accordance with an embodiment of the present application;

fig. 3 is a schematic diagram of smart vehicle-cloud interaction provided in an embodiment of the present application;

FIG. 4 is a schematic diagram of information interaction of a vehicle networking system provided according to an embodiment of the application;

fig. 5 is an exemplary diagram of a dynamic obstacle avoidance apparatus of an autonomous vehicle according to an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a vehicle according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.

The current intelligent driving collision avoidance system based on vehicle-cloud-vehicle communication only has an automatic Braking system (AEB) function based on V2X (vehicle to X, vehicle wireless communication technology), for example, a 5G communication device senses an intelligent driving vehicle carrying the same communication device, senses the motion and static state of a front target vehicle, senses the abnormal behavior of a front vehicle through the communication device, timely takes a safety strategy for a rear vehicle, and timely triggers the AEB function if necessary, so that the safe collision avoidance of the vehicle is realized.

Because a vehicle-cloud-vehicle interactive autonomous avoidance strategy is not provided at present, a vehicle-cloud-vehicle-based interaction scheme can be designed on the basis of considering the configuration of a current vehicle-mounted sensor, the communication capacity of a vehicle and the real-time property of vehicle information transmission, the information of the surrounding environment can be uploaded to a big data cloud end in real time through the real-time perception of the vehicle on the surrounding static environment, other vehicles can utilize the cloud end data to judge the road condition in real time, and although the real-time property of the cloud end data acquisition is limited, certain road information can be acquired under certain conditions, so that the vehicle can make more intelligent behavior decisions.

A dynamic obstacle avoidance method and apparatus for an autonomous vehicle, a vehicle, and a storage medium according to embodiments of the present application will be described below with reference to the accompanying drawings. Specifically, fig. 1 is a schematic flowchart of a dynamic obstacle avoidance method for an autonomous vehicle according to an embodiment of the present disclosure.

As shown in fig. 1, the dynamic obstacle avoidance method for an autonomous vehicle includes the following steps:

in step S101, cloud data of a current travel section of the autonomous vehicle is acquired.

It can be understood that the cloud data of the current driving road section label can be acquired from the big data cloud end, and the cloud data is the road condition data uploaded by other vehicles when passing through the road section.

Specifically, this car and other vehicles all can be with the big data high in the clouds of road perception information upload: when the vehicle recognizes that a road surface jolts on a specific road section and a specific lane, the information is uploaded to the cloud end, the cloud end marks the lane in real time on the high-precision map, and the cloud end is used for receiving the information and making an intelligent decision by other intelligent networking vehicles with cloud end interaction.

In step S102, the actual road condition of the current driving road section is identified according to the cloud data, and whether an abnormal road surface with a road surface flatness smaller than a preset value exists in the current driving road section is determined according to the actual road condition.

The preset value can be specifically calibrated or set and is used for indicating whether the road surface is an abnormal road surface, and the abnormal road surface can include a road surface with characteristics of bump, pothole and the like.

It can be understood that, this application embodiment can be after obtaining the high in the clouds data, can obtain the actual conditions of the highway section of traveling at present through discerning the high in the clouds data to learn the actual road surface condition in front of the vehicle in advance, the intelligent decision-making when being convenient for follow-up autopilot.

In step S103, when there is an abnormal road surface in the current driving road segment, the host vehicle is controlled to execute a corresponding obstacle avoidance action when the actual distance between the host vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold.

The obstacle avoidance threshold may be specifically set or calibrated according to actual conditions, and is not specifically limited.

The vehicle is configured with cloud interaction to receive map information of a cloud in real time to make behavior decision, and when the intelligent vehicle acquires a hollow road surface in a lane in front from the cloud map data, the lane is changed to a safe lane to drive on the premise of meeting a safe lane changing condition.

Controlling the vehicle to execute corresponding obstacle avoidance actions when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value, and the method comprises the following steps: identifying whether a variable lane exists in a current driving road section; and if the lane changing exists, controlling the host vehicle to perform a lane changing action, otherwise, controlling the host vehicle to perform a braking and decelerating action.

It can be understood that the embodiment of the present application can avoid an obstacle through a plurality of ways, such as lane changing in advance or braking in advance, etc., without specific limitations, and taking lane changing in advance or braking in advance as an example, the embodiment of the present application can select lane changing in advance or braking in advance for deceleration according to the actual situation of a driving road section, so as to reduce the safety influence of an abnormal road surface on driving through an optimal strategy, improve the intelligence of an automatic driving decision, and improve the use experience of a user. The advance lane change or advance braking will be explained by the specific embodiments below.

As a possible implementation manner, after controlling the host vehicle to perform the lane change action, the method includes: acquiring the actual position of the abnormal road surface; acquiring an actual image of a road surface corresponding to the actual position; and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

It can be understood that, when the lane change avoids the abnormal road surface, the embodiment of the application can acquire the road surface condition through the sensor of the vehicle, and when the road surface condition is not abnormal, the road surface information is uploaded to the cloud in time, so that other vehicles can realize driving decision according to the latest road surface condition, and invalid automatic driving decision is carried out when other automatic driving vehicles pass through the road section again.

As another possible implementation manner, after controlling the host vehicle to perform the braking deceleration action, the method includes: acquiring the actual position of the abnormal road surface; detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position; and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end so as to update the state of the road surface corresponding to the actual position.

It can be understood that, this application embodiment is when braking detection is through unusual road surface, can detect the road surface condition through the sensor of vehicle to when detecting the road surface condition as an exception, in time upload road surface information to high in the clouds, so that other vehicles realize driving decision according to the newest road surface condition, carry out invalid autopilot decision-making when preventing that other autopilot vehicles from passing this highway section once more.

Specifically, identifying road bump, pothole characteristics: the vehicle provided with the electric power steering system can transmit the torque of a steering wheel of a driver to wheels through the assistance of the motor through the motor and the speed reducing mechanism, so that the driver can easily steer, meanwhile, the steering mechanism can transmit the jolt of the wheels acted on the road surface to the motor through the mechanical mechanism, and the motor simulates the excitation feedback of the road surface to feed the jolt state of the road surface back to the steering wheel. For an automatic driving vehicle, in the automatic driving control process, the road surface bump condition is identified through an electric power steering mechanism, and whether sudden change of the road surface undulation exists is indirectly judged through a hand moment signal fed back from the road surface.

Simultaneously, the vehicle of this application embodiment still has vehicle accurate positioning function: the vehicle-mounted Inertial Measurement Unit (IMU) is used for realizing accurate positioning of the vehicle, when the vehicle runs on a road covered by a high-precision map, lane-level positioning of the vehicle is realized, and the positioning accuracy of the longitudinal direction and the transverse direction of the vehicle is ensured to be smaller than a preset distance, such as 0.2m or 0.1 m.

From this, this application embodiment can be used for intelligent internet vehicle to go when the structured road, through on-vehicle sensor monitoring road condition, like road surface pothole, jolt etc. judge current highway section road condition through direct, indirect mode to upload data to big data high in the clouds in real time, supply all the other intelligent internet vehicles as the reference of traveling, take suitable driving measures when passing through this highway section.

According to the dynamic obstacle avoidance method for the automatic driving vehicle, the front lane condition can be obtained based on the cloud data, and the obstacle avoidance can be performed in advance when an abnormal road surface exists in the front, so that the road surface condition in front of the vehicle can be obtained in advance based on the cloud data, the distance limit of vehicle perception is compensated, a corresponding driving decision is made in advance, the driving safety is improved, meanwhile, the automatic driving decision is more intelligent, and the use experience of a user is improved.

In the following embodiments, a vehicle configured with an electric power steering system and a vehicle networking system is taken as an example, wherein the electric power steering system is used for identifying road bumps, as shown in fig. 2, and a steering mechanism includes components including: the device comprises a steering wheel 1, a rotation angle sensor 2, a torque sensor 3, a speed reducing mechanism 4, a power-assisted motor 5, a motor control unit 6, a vehicle-mounted CAN (Controller Area Network) bus 7, a worm gear mechanism 8 and wheels 9.

Specifically, the steering wheel 1 is used for a driver to directly operate and take over an autonomous vehicle; the corner sensor 2 is used for receiving the actual angle of a steering wheel controlled by a driver, transmitting an angle signal to the motor control unit, and calculating the motor control torque by the motor control unit; the torque sensor 3 is used for judging the magnitude of torque fed back to the motor by a driver or a road surface, and in the embodiment, is mainly used for receiving the torque variation fed back to a Steering wheel by simulating road surface excitation by an Electric Power Steering (EPS); the speed reducing mechanism 4 is used for transmitting the torque of the motor to the steering transmission mechanism to realize the power transmission of a vehicle steering system; the power-assisted motor 5 is used for receiving a torque control instruction of the motor control unit and accurately and efficiently executing a control command of the motor; the motor control unit 6 is used for receiving a steering angle value of a driver, calculating torque required by motor control in real time according to the steering angle and the turning angular speed, controlling the motor to realize steering, receiving excitation fed back to a transmission system from a road surface, simulating the excitation by the motor control unit, feeding the excitation from the road surface back to a steering wheel released by the driver, and enabling the driver to feel a road surface jolt state in time; the vehicle-mounted CAN bus 7 receives a torque value output by the steering system, transmits a torque signal to the automatic driving domain controller in real time, and is used for judging a road bumping state; the worm gear mechanism 8 is used for transmitting the steering torque transmitted by the steering rod to the wheel end;

in addition, before describing the dynamic obstacle avoidance method for the automatic driving vehicle, a data uploading and downloading link between the intelligent vehicle and the big data cloud is described, as shown in fig. 3, specifically as follows:

1. the EPS outputs the moment acted on the steering system by the road surface in real time and transmits the moment to the automatic driving area controller in real time through the CAN bus; when a road surface is hollow and bumpy, the bumpy road surface firstly acts on a vehicle and a suspension system, when a front wheel passes through the bumpy road surface, the front wheel deflects leftwards or rightwards suddenly, the deflection force is transmitted sequentially by the wheel, a front wheel steering rod, a steering worm gear and a steering rod, and finally transmitted to a torque sensor unit, and finally a steering system outputs the torque of a steering wheel in real time by a CAN bus.

2. The vehicle can be accurately positioned on a road section with a high-precision map by depending on an Inertial Measurement Unit (IMU) sensor and a Global Positioning System (GPS); the high-precision map has rich road environment information including static information and dynamic information of roads. Wherein, the static information mainly includes: lane line characteristic, road edge characteristic, road reposition of redundant personnel/information of converging, tunnel, information such as slope, dynamic information includes: whether construction is performed, whether traffic accidents happen or not, whether traffic jam occurs or not and the like.

The automatic driving vehicle needs to use the information of a high-precision map, and also needs to acquire the position of the current position of the vehicle in the high-precision map, in order to realize accurate positioning, information such as a GPS, an IMU, a vehicle-mounted camera, a vehicle-mounted radar and the like is needed, the GPS can be used as a reference for coarse positioning of the vehicle, longitude and latitude information acquired from the GPS can enable the vehicle to realize road-level positioning such as the current road section, the coarse positioning can enable the vehicle to realize positioning with the accuracy of the distance between the transverse direction and the longitudinal direction smaller than a certain distance, such as 1 meter, and lane-level information of the high-precision map needs to be acquired more accurately so as to realize lane-level positioning, a vehicle end receives road environment information output by sensors such as the camera, the radar and the like, such as road boundaries, pavement marks (arrows, lane line types and the like), and ramp diversion/confluence angular point characteristics, semantic characteristics sensed in real time through the vehicle end are matched with semantic information of the high-precision map, the lane-level positioning is realized by a fusion positioning algorithm, and the specific positioning algorithm is a conventional algorithm in the field, so that the redundancy is avoided and the outline is not provided.

3. The automatic driving area controller receives real-time torque change output by the EPS, road surface potholes identified by the camera, road surface abnormity and real-time coordinates output by map positioning, and the road surface state of a specific position of a road is uploaded to the big data cloud after comprehensive processing.

The automatic driving controller needs to receive hand torque information of a steering system, firstly, first-order low-pass filtering is carried out on hand torque signals, and a specific filtering formula is as follows:

EPS_MeasureTorque＝K*EPS_MeasureTorque+(1-K)*EPS_MeasureTorque_Last。

the EPS _ MeasureTorque is a hand moment value at the current moment, K is a first-order low-pass filter coefficient, the EPS _ MeasureTorque _ Last is a hand moment value at the previous moment, and the first-order low-pass filter coefficient value is calibrated according to the actual vehicle state.

The hand moment change value after filtering processing can filter the interference caused by fine bump on the road surface, and can reduce the interference to clutter signals, when the automatic driving vehicle stably runs, the hand moment tends to be stabilized at 0 +/-0.02N m, when the automatic driving vehicle passes through bumpy road surfaces, the hand moment value can change outside a stable range due to left and right change of wheels caused by bumping, along with different bumping degrees, the slope of the hand moment change is different, when the road surfaces bump slowly, the change value of the hand moment slope is smaller, when the road surfaces have pits and bulges, the wheels can deflect rapidly, and at the moment, the slope of the hand moment is larger.

Therefore, whether the road jolt exists or not can be accurately judged through the slope change of the hand moment.

If the automatic driving vehicle runs in a state of rapid change of hand moment, the road condition of the road surface can be indirectly judged to be poor, and the conditions of depression and protrusion can exist, the road fluctuation conditions in the current road and the specific lane can be accurately calculated by combining the lane level positioning output by the positioning module, the calculated road information is output to a T-BOX (Telematics processor), and the road information is uploaded to the cloud end by the T-BOX.

The automatic driving area controller receives cloud data downloaded by the T-BOX, accurately positions the current road and the road condition in front of the current lane by fusing the positioning module, if bumpy or hollow road surfaces exist, when hollow road surfaces exist in front of the lane, the automatic driving area controller can execute a more intelligent decision-making strategy to control the vehicle to change lanes to lanes which are safer and have better road conditions, and the specific lane changing action is executed by controlling the vehicle to automatically change lanes through the EPS.

4. The vehicle-mounted T-BOX is internally provided with a communication module, so that real-time uploading and downloading of a vehicle end and a cloud database can be realized, and the T-BOX receives environment information and coordinate information output by an automatic driving area controller and uploads the environment information and the coordinate information to the cloud database in real time; meanwhile, map data of cloud data can be downloaded through the T-BOX, and the map data can be updated to a vehicle end for an automatic driving area controller to receive and make an intelligent decision.

5. The big data cloud platform has the capacity of updating the map on line at high precision, can realize the on-line marking of the map, and transmits the data to the intelligent vehicle capable of receiving the cloud data.

Based on the vehicle steering system architecture shown in fig. 2 and the intelligent vehicle-big data cloud interaction principle shown in fig. 3, the dynamic obstacle avoidance method for the automatic driving vehicle is shown in fig. 4, and comprises the following specific steps:

1. when a plurality of automatic driving vehicles exist in the road, each automatic driving vehicle can exchange information with the cloud end through the T-BOX, and data is uploaded and downloaded;

2. when a historical vehicle passes through a front pothole road surface on the current road, the previous vehicle judges the road condition in the current lane according to the influence of the road surface on the excitation of the current vehicle, and feeds the current road condition back to the cloud end to update the road environment;

3. when a vehicle passes through the lane again, if the road bumping state is not detected, the lane state of the road section is considered to be good, and the road information also needs to be uploaded to the cloud in real time, so that invalid automatic driving decisions are prevented when the automatic driving vehicle passes through the road section again;

4. when the vehicle automatically drives to pass through the road section, whether the current lane has information such as road surface potholes and jolts or not is judged in real time according to the cloud data, and appropriate behavior decision is made according to the road condition;

5. the big data cloud terminal is a data storage unit and is provided by a relevant unit with map collection qualification, and the data transmission mode and the data transmission protocol related to the embodiment of the application are conventional means in the field, so that redundancy is avoided and repeated description is omitted.

In summary, the embodiment of the application is suitable for a pure electric vehicle and a fuel vehicle equipped with an electric power steering system, is used for identifying the hand moment fed back to the electric power steering from road surface bump, is suitable for a vehicle equipped with a vehicle networking system, is used for interacting the vehicle with cloud data, uploading vehicle data in real time, and receiving the prior information of the cloud to the road surface information, so that the vehicle can make an intelligent decision, thereby realizing more accurate perception of the intelligent networked vehicle to the real-time road information through a vehicle-cloud-vehicle interaction architecture based on the identification of the steering power system to road surface pothole road surfaces, the identification of the camera to road surface pothole road surfaces, a vehicle positioning system based on a GPS, and the design of the vehicle-cloud-vehicle interaction system based on the vehicle networking, and utilizing the prior information to make a safer and more comfortable behavior decision, the driving safety of the vehicle is ensured.

Next, a dynamic obstacle avoidance apparatus for an autonomous vehicle according to an embodiment of the present application will be described with reference to the drawings.

Fig. 5 is a block diagram schematically illustrating a dynamic obstacle avoidance device of an autonomous vehicle according to an embodiment of the present application.

As shown in fig. 5, the dynamic obstacle avoidance apparatus 10 for an autonomous vehicle includes: an acquisition module 100, a recognition module 200 and a control module 300.

The acquisition module 100 is configured to acquire cloud data of a current driving road section of the autonomous vehicle; the identification module 200 is configured to identify an actual road condition of the current driving road section according to the cloud data, and determine whether an abnormal road surface with a road surface flatness smaller than a preset value exists in the current driving road section according to the actual road condition; the control module 300 is configured to, when an abnormal road surface exists in a current driving road segment, control the host vehicle to execute a corresponding obstacle avoidance action when an actual distance between the host vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold.

In the embodiment of the present application, the control module 300 is configured to: identifying whether a variable lane exists in a current driving road section; and if the lane changing exists, controlling the host vehicle to perform the lane changing action, otherwise, controlling the host vehicle to perform the braking and decelerating action.

In this embodiment, the apparatus in this embodiment further includes: the device comprises a first detection module and a second detection module.

The first detection module is used for acquiring the actual position of the abnormal road surface after controlling the vehicle to perform lane changing action; acquiring an actual image of a road surface corresponding to the actual position; and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position. The second detection module is used for acquiring the actual position of the abnormal road surface after controlling the host vehicle to perform lane changing action; detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position; and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end so as to update the state of the road surface corresponding to the actual position.

It should be noted that the foregoing explanation of the embodiment of the dynamic obstacle avoidance method for an autonomous vehicle is also applicable to the dynamic obstacle avoidance apparatus for an autonomous vehicle in this embodiment, and is not repeated herein.

According to the dynamic obstacle avoidance device of the automatic driving vehicle, the front lane condition can be acquired based on the cloud data, the obstacle is avoided in advance when an abnormal road surface exists in the front, the road condition in the front of the vehicle can be acquired in advance based on the cloud data, the distance limitation of vehicle perception is compensated, a corresponding driving decision is made in advance, the driving safety is improved, the automatic driving decision is made to be more intelligent, and the use experience of a user is improved.

Fig. 6 is a schematic structural diagram of a vehicle according to an embodiment of the present application. The vehicle may include:

a memory 601, a processor 602, and a computer program stored on the memory 601 and executable on the processor 602.

The processor 602, when executing the program, implements the dynamic obstacle avoidance method for an autonomous vehicle provided in the above embodiments.

Further, the vehicle further includes:

a communication interface 603 for communicating between the memory 601 and the processor 602.

The memory 601 is used for storing computer programs that can be run on the processor 602.

The Memory 601 may include a high-speed RAM (Random Access Memory) Memory, and may also include a non-volatile Memory, such as at least one disk Memory.

If the memory 601, the processor 602 and the communication interface 603 are implemented independently, the communication interface 603, the memory 601 and the processor 602 may be connected to each other through a bus and perform communication with each other. The bus may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component interconnect) bus, an EISA (Extended Industry Standard Architecture) 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 thick line is shown in FIG. 6, but that does not indicate only one bus or one type of bus.

Optionally, in a specific implementation, if the memory 601, the processor 602, and the communication interface 603 are integrated into a chip, the memory 601, the processor 602, and the communication interface 603 may complete mutual communication through an internal interface.

The processor 602 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present Application.

Embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the above dynamic obstacle avoidance method for an autonomous vehicle.

In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

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, "N" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

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 N executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application 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 implementing the embodiments of the present application.

It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. 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, a field programmable gate array, or the like.

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

Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.

Claims (10)

1. A dynamic obstacle avoidance method of an automatic driving vehicle is characterized by comprising the following steps:

acquiring cloud data of a current driving road section of an automatic driving vehicle;

identifying the actual road condition of the current driving road section according to the cloud data, and judging whether an abnormal road surface with the road surface evenness smaller than a preset value exists in the current driving road section according to the actual road condition;

and when the abnormal road surface exists in the current running road section, controlling the vehicle to execute corresponding obstacle avoidance action when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value.

2. The method of claim 1, wherein the controlling the host vehicle to perform a corresponding obstacle avoidance action when the actual distance from the host vehicle to the abnormal road surface is less than an obstacle avoidance threshold comprises:

identifying whether a variable lane exists in the current driving road section;

and if the lane changing exists, controlling the host vehicle to perform a lane changing action, otherwise, controlling the host vehicle to perform a braking and decelerating action.

3. The method of claim 2, after controlling the host-vehicle to perform a lane-change action, comprising:

acquiring the actual position of the abnormal road surface;

acquiring an actual image of the road surface corresponding to the actual position;

and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

4. The method of claim 2, after controlling the host-vehicle to perform a braking deceleration action, comprising:

acquiring the actual position of the abnormal road surface;

detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position;

and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end so as to update the state of the road surface corresponding to the actual position.

5. A dynamic obstacle avoidance device for an autonomous vehicle, comprising:

the acquisition module is used for acquiring cloud data of a current driving road section of the automatic driving vehicle;

the identification module is used for identifying the actual road condition of the current driving road section according to the cloud data and judging whether an abnormal road surface with the road surface evenness smaller than a preset value exists in the current driving road section according to the actual road condition;

and the control module is used for controlling the vehicle to execute corresponding obstacle avoidance actions when the actual distance between the vehicle and the abnormal road surface is smaller than an obstacle avoidance threshold value when the abnormal road surface exists on the current running road section.

6. The apparatus of claim 5, wherein the control module is configured to:

identifying whether a variable lane exists in the current driving road section;

and if the lane changing exists, controlling the host vehicle to perform a lane changing action, otherwise, controlling the host vehicle to perform a braking and decelerating action.

7. The method of claim 6, further comprising:

a first detection module for acquiring an actual position of the abnormal road surface after controlling the host vehicle to perform a lane change action; acquiring an actual image of the road surface corresponding to the actual position; and identifying the actual image to obtain the actual state of the road surface corresponding to the actual position, and if the actual state is a non-abnormal state, uploading the actual state of the road surface corresponding to the actual position to a cloud end to update the state of the road surface corresponding to the actual position.

8. The apparatus of claim 6, further comprising:

the second detection module is used for acquiring the actual position of the abnormal road surface after controlling the host vehicle to execute lane changing action; detecting a plurality of hand moment values of the vehicle passing through a road surface corresponding to the actual position; and calculating a hand moment slope change value according to the hand moment values, judging that the actual state of the road surface corresponding to the actual position is a non-abnormal state when the hand moment slope change value is smaller than an abnormal threshold value, and uploading the actual state of the road surface corresponding to the actual position to a cloud end so as to update the state of the road surface corresponding to the actual position.

9. A vehicle, characterized by comprising: memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method of dynamic obstacle avoidance for an autonomous vehicle as claimed in any of claims 1-4.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program is executed by a processor for implementing a dynamic obstacle avoidance method for an autonomous vehicle as claimed in any of claims 1-4.

Images