CN107678306A - Dynamic scene information is recorded and emulation back method, device, equipment and medium - Google Patents

Dynamic scene information is recorded and emulation back method, device, equipment and medium Download PDF

Info

Publication number
CN107678306A
CN107678306A CN201710930672.6A CN201710930672A CN107678306A CN 107678306 A CN107678306 A CN 107678306A CN 201710930672 A CN201710930672 A CN 201710930672A CN 107678306 A CN107678306 A CN 107678306A
Authority
CN
China
Prior art keywords
scene
dynamic
vehicle
planning control
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201710930672.6A
Other languages
Chinese (zh)
Other versions
CN107678306B (en
Inventor
耿秀军
周小成
徐澍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Uisee Shanghai Automotive Technologies Ltd
Original Assignee
Uisee Shanghai Automotive Technologies Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Uisee Shanghai Automotive Technologies Ltd filed Critical Uisee Shanghai Automotive Technologies Ltd
Priority to CN201710930672.6A priority Critical patent/CN107678306B/en
Publication of CN107678306A publication Critical patent/CN107678306A/en
Application granted granted Critical
Publication of CN107678306B publication Critical patent/CN107678306B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B17/00Systems involving the use of models or simulators of said systems
    • G05B17/02Systems involving the use of models or simulators of said systems electric

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Traffic Control Systems (AREA)

Abstract

Back method and device, this method are recorded and emulated the present invention relates to dynamic scene information to be included:Sensing data is collected, the conversion from local coordinate to world coordinates is carried out to sensing data, to generate absolute location information;Model and scene are established according to absolute location information, to realize the reduction of road environment, and the conversion from world coordinates to local coordinate is carried out to sensing data, to generate relative position information, and according to relative position information come the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene;According to planning control information and relative position information come planning control vehicle, to verify improved planning control algorithm.The device includes:Pretreatment module, emulation module and planning control module.The present invention can form a fixed regression test scene, and the scene is not only stablized and can reappear failure problems at that time, even if planning control algorithm, which changes, nor affects on the reproduction of failure problems and the reduction of scene.

Description

Dynamic scene information is recorded and emulation back method, device, equipment and medium
Technical field
The application is related to emulation and the checking field of intelligent driving vehicle, and records and emulate more particularly, to one kind The method and device of playback.
Background technology
In order to strengthen the stability of intelligent driving vehicle, it is necessary to substantial amounts of real vehicle drive test.At present, when finding to ask in drive test After topic, drive test data can be transferred to research staff to handle by tester.The data that research staff is recorded by analyzing in drive test point Analysis problem is simultaneously found after the basic reason of problem, it is necessary to tester returns the scene to go wrong at that time is tested, and is seen and is worked as When the problem of whether be resolved.
Problems be present in prior art:
1. prior art only gathers and recorded position of the object relative to sensor.Therefore, once control algolithm occurs Change, the data of relative position will change, and dynamic scene will change during playback.According to the relative position of object The scene of reduction is only capable of showing the course of work of vehicle under former control algolithm.Therefore, when research and development are to avoid accident or raising vehicle Performance and improve planning control algorithm and want verify improved planning control algorithm when, be merely capable of by road actual measurement come Realize, waste a large amount of man power and materials.
2. because the road environment of reality often changes.Therefore, in by actual environment many uncertain factors shadow Ring, it is difficult to reappear the scene the same with during vehicle generation problem.In road actual measurement, calculated using improved control The method and failure problems of vehicle no longer occur, can not also ensure really solve the failure problems.
3. after control algolithm repeatedly updates, new control algolithm may cause the partial function of old control algolithm to fail Or it is affected.Therefore, even if solving original failure problems under old control algolithm, can not ensure in new control calculation Under method, whether the failure problems can be triggered again.So, test is required for entering for all failure problems occurred every time Survey on trade road.The environment of road actual measurement, easy error of not only wasting time and energy are built repeatedly.Also, carry out the ring of road actual measurement Border is very unstable, easily causes test result distortion so that the new defect of the planning control algorithm under redaction is difficult to be found And old defect is difficult to really avoid, cause serious traffic accident occurs on vehicle behind road.
Therefore, existing emulation and verification method existing defects.
The content of the invention
In order to solve the above-mentioned technical problem, it is proposed that the application.Embodiments herein provides a kind of dynamic scene ring The record of border perception information and the method and device of emulation playback, it can form a fixed regression test scene, this Scape is not only stablized and can reappear failure problems at that time, even if planning control algorithm, which changes, nor affects on failure problems Reproduction and the reduction of scene, not only can ensure that the authentic and valid of test result, and can greatly reduce the cost of test.
The invention provides a kind of recording of dynamic scene information and emulation back method, including:
Sensing data is collected, the conversion from local coordinate to world coordinates is carried out to sensing data, to generate absolute position Information, wherein, absolute location information is positional information of the target under global coordinate system;
Scene is established according to absolute location information, to realize the reduction of road environment, and sensing data is carried out from World coordinates to local coordinate conversion, to generate relative position information, and according to relative position information come analogue simulation car Traveling process, to realize the reduction of user's Driving Scene, wherein, relative position information be target under local coordinate system Positional information;
According to planning control information and relative position information come planning control vehicle, to verify that improved planning control is calculated Method, wherein, planning control information includes the artificial means of planning control and the planning control of Automatic Optimal vehicle for changing vehicle Means.
It is preferred that establishing scene includes:
Dynamic Data Processing and static data processing are carried out to the sensing data under global coordinate system, to generate dynamic model And static models;
Dynamic model and static models are combined, to generate dynamic scene and static scene;
Dynamic scene and static scene are combined, to realize the reduction of road environment.
It is preferred that establishing scene also includes:The dynamic scene and/or static scene of generation are stored, with It is provided with being used directly to combine scenes during post-simulation, to realize the reduction of road environment.
It is preferred that dynamic scene and static scene are combined including:
Dynamic scene is merged with static scene according to coordinate information, to realize the reduction of road environment.
It is preferred that Dynamic Data Processing and static data processing are carried out to sensing data to be included:
It is dynamic for identifying to add dynamic labels and static labels respectively to the dynamic data in sensing data and static data State data and static data.
It is preferred that the traveling process of analogue simulation vehicle includes:
The sensing data after planning control information and conversion from world coordinates to local coordinate is received, and for sensing number Dynamic data in is emulated, and carrys out the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene.
It is preferred that conversion from local coordinate to world coordinates or world coordinates are carried out to sensing data to part Transformation of coordinates includes:
It is online to change sensing data in real time and/or change sensing data offline.
It is preferred that relative position information includes real-time vehicle body positional information and obstacle under first person Object location information;
Absolute location information includes real-time the vehicle body positional information and obstacle position information under third person.
According to a further aspect of the invention, a kind of recording of dynamic scene information and emulation playback reproducer, including:
Pretreatment module, for carrying out the conversion from local coordinate to world coordinates to sensing data, to generate absolute position Confidence ceases;
Emulation module, for establishing scene according to absolute location information, to realize the reduction of road environment, and to passing Conversion of the data progress from world coordinates to local coordinate is felt, to generate relative position information, and according to relative position information Carry out the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene;
Planning control module, for according to planning control information and relative position information come planning control vehicle, with checking Improved planning control algorithm, wherein, planning control information includes the means of the artificial planning control for changing vehicle and automatic excellent Change the means of the planning control of vehicle.
It is preferred that the pretreatment module includes:
Acquisition module, for collecting sensing data.
It is preferred that the pretreatment module includes:
Coordinate transformation module, it is absolute to generate for carrying out the conversion from local coordinate to world coordinates to sensing data Positional information.
It is preferred that the pretreatment module includes:
Data processing module, for entering action to the dynamic data in the sensing data under global coordinate system and static data State data processing and static data processing, to generate dynamic model and static models.
It is preferred that the emulation module includes:
Scenario building unit, carried out for receiving dynamic model and static models, and to dynamic model and static models Combination, to generate dynamic scene and static scene, and is combined to dynamic scene and static scene, to realize road environment Reduction.
It is preferred that recording and the emulation playback reproducer of a kind of dynamic scene information, in addition to:
Memory module, enter for receiving dynamic scene and/or static scene, and to dynamic scene and/or static scene Row storage, for be used directly to combine scenes during post-simulation, to realize the reduction of road environment.
It is preferred that the emulation module includes:
Coordinate transformation unit, it is relative to generate for carrying out the conversion from world coordinates to local coordinate to sensing data Positional information.
It is preferred that the emulation module includes:
Dynamic simulation unit, for receiving the sensing after planning control information and conversion from world coordinates to local coordinate Data, and emulated for the dynamic data in sensing data, carry out the traveling process of analogue simulation vehicle, to realize that user drives Sail the reduction of scene.
It is preferred that the planning control module includes:
Unmanned Systems, for according to planning control information and relative position information come planning control vehicle, with checking Improved planning control algorithm.
According to the another aspect of the application, there is provided a kind of electronic equipment, including:Processor;Memory;And it is stored in Computer program instructions in memory, computer program instructions cause computing device is above-mentioned to move when being run by processor The method of recording and the emulation playback of state scene information.
According to the another aspect of the application, there is provided a kind of computer-readable recording medium, be stored thereon with computer journey Sequence instruct, computer program instructions cause when being run by processor the above-mentioned dynamic scene information of computing device recording and The method for emulating playback.
Compared with prior art, the recording of dynamic scene information according to embodiments of the present invention and emulation back method and dress Put, it can form a fixed regression test scene, and the scene is not only stablized and can reappear failure problems at that time, i.e., Planning control algorithm is changed and nor affect on the reproduction of failure problems and the reduction of scene, not only can ensure that test result It is authentic and valid, and can greatly reduce the cost of test.The present invention can generate according to barrier relative to the position of sensor Absolute location information, and emulate barrier using absolute location information and be combined into scene, to ensure the road ring of emulation Border can reduce real road environment completely.The present invention, can be by the absolute position in the scene of reduction in controlling planning Information is converted to relative position information, and carries out planning control to vehicle by receiving default planning control information, comes The process of vehicle drive under real reduction first person.By the true reduction of road environment and vehicle operation, To verify that certain planning control algorithm whether there is problem.
In addition, the road environment of present invention energy storage emulation generation, and can be emulated by combining different road environments Go out the abundant road environment of species, to verify planning control algorithm.
The present invention needs only to collect data during vehicle traveling, just can truly restore road environment and vehicle drive Process, the scene of reduction is stable and can preserve, test data accurate and effective.Therefore, the present invention can pass through emulation completely To realize the checking of planning control algorithm, without carrying out road actual measurement, the error and cost of test can be greatly reduced.The present invention The scene of emulation, it can effectively verify planning control algorithm.The scene of high reduction degree can cause tester to be very easy to find rule Draw the defects of control algolithm is present and improved.Moreover, improved planning control algorithm can be imported the present invention to be imitated Very, tester can verify that improved planning control algorithm whether there is defect whenever and wherever possible, can so greatly improve planning control The robustness of algorithm processed.The present invention can also generate more test scenes by the road environment of combining simulation, each to meet The needs of class testing.
Brief description of the drawings
Fig. 1 illustrates the recording according to the dynamic scene information of the embodiment of the present application and the schematic frame of emulation playback reproducer Figure.
Fig. 2 illustrates the schematic block diagram of the pretreatment module according to the embodiment of the present application.
Fig. 3 illustrates the schematic block diagram of the emulation module according to the embodiment of the present application.
Fig. 4 illustrates the schematic flow sheet collected and pre-processed according to the sensing data of the embodiment of the present application.
Fig. 5 illustrates the schematic flow sheet played back according to the emulation of the embodiment of the present application.
Fig. 6 illustrates the step schematic diagram of recording and the emulation playback according to the dynamic scene information of the embodiment of the present application.
Embodiment
Below, the example embodiment according to the application will be described in detail by referring to the drawings.Obviously, described embodiment is only Only it is the part of the embodiment of the application, rather than the whole embodiments of the application, it should be appreciated that the application is not by described herein The limitation of example embodiment.
Application general introduction
As described above, there is following defect in existing emulation with verification method.
1. prior art only gathers and recorded position of the object relative to sensor.Therefore, once control algolithm occurs Change, the data of relative position will change, and dynamic scene will change during playback.According to the relative position of object The scene of reduction is only capable of showing the course of work of vehicle under former control algolithm.Therefore, when research and development are to avoid accident or raising vehicle Performance and improve planning control algorithm and want verify improved planning control algorithm when, be merely capable of by road actual measurement come Realize, waste a large amount of man power and materials.
2. because the road environment of reality often changes.Therefore, in by actual environment many uncertain factors shadow Ring, it is difficult to reappear the scene the same with during vehicle generation problem.In road actual measurement, calculated using improved control The method and failure problems of vehicle no longer occur, can not also ensure really solve the failure problems.
3. after control algolithm repeatedly updates, new control algolithm may cause the partial function of old control algolithm to fail Or it is affected.Therefore, even if solving original failure problems under old control algolithm, can not ensure in new control calculation Under method, whether the failure problems can be triggered again.So, test is required for entering for all failure problems occurred every time Survey on trade road.The environment of road actual measurement, easy error of not only wasting time and energy are built repeatedly.Also, carry out the ring of road actual measurement Border is very unstable, easily causes test result distortion so that the new defect of the planning control algorithm under redaction is difficult to be found And old defect is difficult to really avoid, cause serious traffic accident occurs on vehicle behind road.
For the technical problem, the basic conception of the application is (from office by the sensing data of collection by coordinate direct transform Portion's coordinate system transformation is to global coordinate system), and then the absolute position of object is obtained to reduce road environment.Again by after direct transform Sensing data obtains the relative position of object by inverse transformation (transforming to local coordinate system from global coordinate system), to go back The driving procedure of former vehicle, and planning control is carried out to vehicle.
On the one hand, the present invention only records object in the prior art by the coordinate direct transform to sensing data to solve Relative position when, when planning control algorithm changes, because new planning control algorithm can cause the relative position between object Put and change, and can not reflect what new planning control algorithm was brought by the initial data of collection to build simulating scenes Change, and then cause to emulate the problem of failing.Also, the stability and reduction degree of the scene that the present invention is built are high, can reappear Be out of order problem.The scene of the present invention can preserve to verify that the planning control algorithm of different editions whether there is above-mentioned failure Problem, without being surveyed by road.
On the other hand, by carrying out inverse transformation to sensing data to emulate the traveling process of vehicle so that the phase between object Determination to position, independent of planning control algorithm.The real-time driving data such as the relative position in analogue system, it is to pass through emulation The planning control information of system real-time reception come start or close vehicle traveling in intelligent operation (planning control information is producer Generated using planning control algorithm, producer generates planning control information by inputting planning control algorithm, and planning is controlled Information processed is input in the analogue system of the present invention, it becomes possible to goes out the car of a certain type come real analogue simulation by the system Intelligence system).Also, specific operation of the sensing data energy simulating vehicle collected in specific road environment is to the shadow of driving Ring.Therefore, if the system receive in real time the operational order of planning control information or tester just can automatically by Correspondingly operation is converted to relative position parameter etc..Also, these parameters can be shown in analogue system in real time.
Therefore, the present invention can overcome the recorded broadcast that is merely capable of of prior art, and the problem of planning control algorithm can not be verified. The present invention can reappear the scene the same with during vehicle generation problem, reappear failure problems, and checking in any type Number vehicle on, during using arbitrary planning control algorithm, if existing defects, surveyed without road.Also, the present invention is raw Into scene can preserve whenever and wherever possible, miscellaneous combination is carried out, for any one planning control algorithm of Future authentication When use.
The concrete principle of the present invention is as follows:
Sensing data is transformed under global coordinate system, generates the absolute location information of each barrier in road environment. Scene is established by absolute location information, to reduce real road environment, wherein, road environment is, for example, pedestrian, Qi Tache , trees, street lamp and landform etc..Vehicle is incorporated in the road environment of emulation, the emulation playback to vehicle operation is By calculating the relative position of vehicle and road environment come what is realized, the world coordinates of barrier and vehicle is converted in real time Local coordinate.
The main purpose of the emulation playback of the present invention is to verify planning control algorithm by emulating reduction scene.Specifically Ground, can be by generating the scene of road environment and vehicle drive come verification algorithm.The scene of road environment can pass through combination Model, and model is put into scene map composition scene according to absolute location coordinates, then different scene maps is combined Reduce the road environment of an entirety.The information of road surface of scene map then can be by modeling to generate to sensing data.Due to The emulation playback of the present invention is to build Scene realization by the sensing data (for example, data that various sensors gather) of collection , thus can truly restore scene at that time.
The present invention proposes method, apparatus, the electronics that a kind of record of dynamic scene environment sensing information plays back with emulation Equipment and computer-readable recording medium, it can play the maximum effect for the real data collected in drive test, pole as far as possible The testing cost of big reduction intelligent driving, and truly emulation restores scene at that time to verify planning control algorithm.
After the general principle of the application is described, carry out the specific various non-limits for introducing the application below with reference to the accompanying drawings Property embodiment processed.
Exemplary means
Below, the recording and emulation of the dynamic scene information according to the embodiment of the present application are described with reference to figure 1, Fig. 2 and Fig. 3 Playback reproducer.
Fig. 1 illustrates the recording according to the dynamic scene information of the embodiment of the present application and the schematic frame of emulation playback reproducer Figure.As shown in figure 1, the recording of dynamic scene information and emulation playback reproducer can include:
Pretreatment module 101, it is absolute to generate for carrying out the conversion from local coordinate to world coordinates to sensing data Positional information, wherein described absolute location information refers to positional information of the target under global coordinate system, i.e., with earth point The positional information of earth surface object, such as UTM coordinate systems are described for coordinate origin.
Emulation module 102, for establishing scene according to absolute location information, to realize the reduction of road environment, and Conversion from world coordinates to local coordinate is carried out to sensing data, to generate relative position information, and according to relative position Information carrys out the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene, wherein, described relative position information Refer to positional information of the target under local coordinate system.In certain embodiments, described local coordinate system refers to vehicle coordinate System, i.e., centered on vehicle, the horizontal and vertical coordinate system for reference axis of vehicle.
Planning control module 103, for according to planning control information and relative position information come planning control vehicle, to test Improved planning control algorithm is demonstrate,proved, wherein, planning control information is including the artificially means of the planning control of change vehicle and automatically Optimize the means of the planning control of vehicle.
Pretreatment module 101 is used to pre-process the sensing data that vehicle collects, wherein, pretreatment is included from office Portion's coordinate to world coordinates conversion, to generate absolute location information.Sensing data can include many kinds of parameters, such as light is strong The operating parameter etc. of degree, map size, topography and geomorphology, road environment, vehicle.Light intensity, map size, topography and geomorphology etc. Parameter is mainly acquired on the spot, such as hills, mountain pass, residential block etc., gathered data can pass through camera, the gradient Measuring instrument, sensitometer etc..Sensing data can be acquired by sensor.Sensor can include binocular camera, laser Radar, ultrasonic sensor etc..Sensing data can be obtained by multi collect.
The data of camera shooting typically only have recorded position of the object relative to camera.Because camera is usually to pacify On vehicle, so in order to really determine that the position of object has in real time be transformed into the positional information of the object of collection Being recorded under global coordinate system, the absolute location information so recorded would not be influenceed by cam movement, therefore, Absolute location information is fixed, is not influenceed by the course and planning control algorithm of vehicle.
In the present embodiment, in order to generate scene, it is necessary to according to absolute location information generation model, the process of generation model Both it can be realized, can also be realized by emulation module 102 by pretreatment module 101.
Emulation module 102 can be according to the sensing data under global coordinate system or absolute location information come built-up pattern, with life Into scene, combine scenes generate road environment, and according to relative position information come the traveling process of analogue simulation vehicle.
In the present embodiment, generation road environment can include dyspoiesis thing, and generation terrain environment data etc..Road Mainly the parameter such as the position including barrier, size, shape, color, these parameters are used to build in scene road environmental data Barrier.Barrier can be static state or be dynamic.Barrier can be arranged according to absolute location information, with structure Into scene map.Road environment can also include the data such as light intensity, scene map size, topography and geomorphology, with emulation Simulate the terrain environment at scene.Barrier, according to the absolute position of respective objects, can be arranged in tool according to absolute location information Have on the scene map of a certain terrain environment, to generate the high test scene of reduction degree.Barrier and the mistake of terrain environment modeling Journey can use existing VR technologies.The collection of road environment data depends on the shooting of high speed camera.
The traveling process of analogue simulation vehicle, specifically, can be according to the real-time position of each object under global coordinate system Put calculate the vehicle of emulation under steam around which direction on and more have barrier at a distance (barrier can be moved Dynamic).Analogue system automatically extracts the operation of vehicle from planning control information.The emulation of operation to vehicle can pass through receipts Data when collecting vehicle drive obtain.For example, to be stepped on the gas and turn the position of the operation to vehicle such as steering wheel and vehicle The data such as direction emulation, can be stepped on the gas and deflection data during steering wheel rotation by gathering the actual vehicle of the model Obtained with displacement data, can so reach very real simulated effect.Emulation module 102 calculates partially according to sensing data The influence to vehicle and environment such as revolution evidence and displacement data.
During planning control is carried out to vehicle according to planning control information, the energy real-time update of emulation module 102 simultaneously shows Vehicle and the state with peripheral obstacle are shown, the traveling process of vehicle is gone out come truly analogue simulation.
Therefore, tester can be observed the traveling process of vehicle in the analogue system of the present invention, to test After planning control algorithm changes, vehicle whether can avoiding obstacles, and then improved planning control algorithm is verified.
In the present invention, local coordinate system can be vehicle axis system.Vehicle axis system be using vehicle as coordinate origin, The horizontal and vertical of vehicle is reference axis.Global coordinate system describes earth surface object so that earth point is coordinate origin Positional information.Global coordinate system can include UTM coordinate systems.
Specifically, in certain embodiments, emulation module 102 can receive concrete operations information matches what is collected in advance Every sensing data during real steering vectors is also, real-time by effect come specific effect caused by describing a certain operation of vehicle Renewal display in the present system, and then truly simulates the vehicle of disposable type state in motion and to peripheral obstacle Role and influence etc..So, even if planning control algorithm changes, the change of algorithm also can be truly reflected in this In system.Tester can be by observing the system, intuitively to be verified to planning control algorithm.
Emulation module 102 describes the process of the specific effect of a certain operation of vehicle by sensing data, may be referred to existing In some VR technologies, the process of dynamic model and scene interactivity, the present invention is no longer repeated this.In virtual system, behaviour Existing VR technologies can also be referred to by making the mode of dynamic model (such as vehicle), provide many first persons in the prior art Realize that the modeling of interior environment, and operation model drive vehicle, and the outer scenery of vehicle window under visual angle by VR technologies The method of the realizations such as change, the present invention are no longer repeated this.
In certain embodiments, pretreatment module 101 is to sensing data direct transform, and does not differentiate between dynamic data and static number According to.Emulation module 102 receives the sensing data of direct transform (conversion from local coordinate to world coordinates).Pretreatment module 101 Or emulation module 102 according to the sensing data after direct transform establishes model (model can also not differentiate between dynamic model and static state Model, and all as dynamic model processing), emulation module 102 builds scene (this according to model and absolute location information Scape can also not differentiate between dynamic scene and static scene, and all as dynamic scene processing), and combine scenes and model construction Road environment, the traveling process that the planning control information that planning control module 103 is sent carrys out simulating vehicle is received, and then verify rule Draw control algolithm.
Planning control module 103 can control the mobile route and drive manner of vehicle.It is provided with planning control module 103 The planning control algorithm of manufacturer's input.Tester corrects the mobile route of vehicle and driving by amending plans control algolithm Mode.Emulation module 102 can parse the operational order of vehicle, the sensing number of matching real vehicle collection from planning control information According to, sensing data is updated, and by effect caused by sensing data including in analogue system.Planning control module 103 is according to imitative The sensing data that true module 102 updates analyze vehicle will how controlling planning, and export planning control information (planning control believed Breath is describing the operating procedure of vehicle planning control) to emulation module 102.Planning control module 103 is used to realize to emulation The planning control of vehicle in module 102.The sensing data of renewal is constantly transmitted to planning control module 103 by emulation module 102 Carry out the change of simulated roadway environment.The real-time reception of planning control module 103 describes the sensing data of road environment change and processing, Do not stop the course and drive manner of amendment vehicle according to planning control algorithm.Planning control module 103 is controlled according to planning Algorithm processed can be just analyzed road environment automatically, and selects suitable drive manner, and traffic route is carried out certainly Dynamic amendment.How planning control algorithm, and the algorithm carry out planning control to vehicle and may be referred to existing planning control skill Art, the technology are widely used in the unpiloted system of vehicle, and the present invention repeats no more to this.Production firm can be allowed to provide The ready-made intelligent driving system of vehicle planning control or Unmanned Systems.The intelligent driving system or Unmanned Systems make For planning control module 103, to realize the recording of the present invention and emulation playback reproducer.
Fig. 2 illustrates the schematic block diagram of the pretreatment module 101 according to the embodiment of the present application.As shown in Fig. 2 according to this The pretreatment module 101 of application embodiment can include:Acquisition module 201, for collecting sensing data.Data processing module 202, for carrying out Dynamic Data Processing and static state to the dynamic data in the sensing data under global coordinate system and static data Data processing, to generate dynamic model and static models.Coordinate transformation module 203, for carrying out sensing data from local seat The conversion of world coordinates is marked, to generate absolute location information.
In certain embodiments, pretreatment module 101 carries out dynamic data identification to data and static data identifies, and builds Vertical dynamic model and static models.
Barrier generally comprises pedestrian, animal, vehicle, rock, guardrail, trees, lamp stand etc. on road.Pedestrian, animal, car It is real time kinematics to wait barrier, so it is fixed in real time typically to determine that positional information needs to carry out these dynamic barriers Position.Dynamic data not only refers to the data of position real-time change, can also recognize when the other specification of jobbie changes To be dynamic data, such as shape, size, color etc..When gathering dynamic data as the barrier of above-mentioned motion, it is necessary to The continuous coordinate direct transform being continuously shot with process chip to sensing data dependent on high speed camera.Moved as record During state data, it usually needs stored in chronological order, the motion of the object moved at that time could be so reduced in emulation State.And the barrier such as rock, guardrail, trees, lamp stand, typically actionless, the parameter of these objects can be described as Static data.Such stationary body avoids the need for captured in real-time and real-time change data, can also be after data have been collected Its coordinate is calculated again, the parameters such as positional information, reconvert to global coordinate system can be gathered with direct labor.If collect simultaneously To dynamic data and static data, coordinate direct transform can also be carried out together, and this avoid real-time distinguish to move static data It is cumbersome.Dynamic data and static data are entered by the identification method of existing dynamic static data again after collection finishes data Row identification.
Specifically, in certain embodiments, processing of the present invention to dynamic data and static data, can both locate together Reason, can also individually be handled.Identification of the present invention to dynamic data and static data can pass through the dynamic static number of prior art According to recognizer realize, for example, specifically, dynamic static data can be identified by prior art, and to moving static data Stamp data type label respectively, output data type label supplies the analogue system of the present invention to identify dynamic data and quiet State data.Data type label includes dynamic labels (expression dynamic data) and static labels (expression static data), such as with 1 Position identifies dynamic data and static data, such as 0 represent static data, and 1 represents dynamic data.Except data type label, Timestamp can also be stamped to sensing data, wherein, timestamp is used for the time for showing sensing data collection, in order to road The process of environment and vehicle traveling carries out dynamic simulation.
Sensing data can include static data and dynamic data.Scene can also include dynamic scene and static scene. It is exactly static data that parameter does not change with the time.It is exactly dynamic data that parameter, which changes over time,.Scene does not change over time It is static scene.It is dynamic scene that scene, which changes over time,.The change of parameter generally refers to the change under certain coordinate system.Example If global coordinate system can be to refer to position changes in global coordinate system sensing data.Pretreatment can include to quiet State data processing and Dynamic Data Processing, wherein, static data processing and Dynamic Data Processing can also include to sensing data Carry out the process of dynamic data modeling and static data modeling.Because dynamic data modeling is real-time, and static data modeling is real When property is weak.Therefore, dynamic data modeling and static data modeling can both be uniformly processed, and can also individually handle.Individually processing Benefit be that operand is greatly reduced, improve system effectiveness.The reason for dynamic model and static models can individually be established be because For:Dynamic data modeling typically confirms the absolute position of object by real-time coordinates direct transform.Static data modeling is then general Real-time modeling set is not needed, and needs only to model in map rejuvenation.
In the present embodiment, sensing data and transmission sensing data are collected to data processing module by acquisition module 201 202 are used to carry out sensing data Dynamic Data Processing and static data processing, and by Dynamic Data Processing and static data Sensing data after reason is transmitted to coordinate transformation module 203 and (converted with carrying out direct transform to the sensing data from local coordinate system To global coordinate system).Coordinate transformation module 203 transmits the sensing data after direct transform to data processing module 202.At data Reason module 202 is modeled according to the sensing data after direct transform.
The present embodiment is it is further preferred that the sensor of acquisition module 201 can include:Vision sensor, laser radar And/or ultrasonic sensor, and other sensors available for pilotless automobile.Vision sensor can be taken the photograph including depth As head and binocular camera shooting are first-class.The sensing data that sensing data obtains for one or many collections.Sensing data can include quiet State data and dynamic data.Static data refers in gathered data, the data not changed under certain coordinate system.It is dynamic State data are opposite with static data.
The present embodiment it is further preferred that data processing module 202 can to sensing data carry out Dynamic Data Processing and Static data processing, and dynamic data and static data are generated, and modeled according to dynamic data core static data.For example, Data processing module 202 can receive absolute location information, and dynamic to identify according to the recognition methods of existing dynamic static data State data and static data.Data processing module 202 individually can also be modeled or modeled together to dynamic data and static data.
Static data processing includes static models structure, such as establishes test scene model according to static data.Dynamic field Scape processing includes dynamic model and built, wherein, dynamic model being capable of the motion rail that changes over time of the accurate description dynamic object Mark.
In certain embodiments, individually the movement locus of the dynamic object in sensing data can be protected sequentially in time Deposit, to generate dynamic model.So in emulation, the sensing data of corresponding time can be directly extracted to generate dynamic scene. Because static data will not change over time, dynamic data and static data can be individually handled, to reduce data processing Amount.
In certain embodiments, emulation module 102 can receive the sensing data after the direct transform of pretreatment module 101, with And receive dynamic model and static models that pretreatment module 101 is built.Emulation module 102 is according to the sensing number after direct transform Dynamic scene and static scene are built according to, dynamic model and static models, and combine scenes and model build road environment. Emulation module 102 can also receive the traveling process that the planning control information that planning control module 103 is sent carrys out simulating vehicle, enter And verify planning control algorithm.
Static scene will not change over time.Dynamic scene can change over time.Therefore, dynamic scene can be individually created And static scene, to reduce operand.Because static scene only needs processing once, and need not handle in real time.In some embodiments In, static scene structure can by the way that the cloud data of laser radar or depth camera is inputed into emulation module 102, In emulation module 102 according to a cloud scan come static-obstacle thing three-dimensional information and positional information establish three-dimensional object model (such as trees, street lamp etc.).And the emulation of dynamic scene is then needed in real time to coordinate inversion, and renewal show object relative to The position of vehicle.
In certain embodiments, can be by the way that dynamic model and static models combination be generated into scene, the scene can be with Including the pedestrian for example on certain road and the dynamic model of vehicle, and the static models such as flowers, plants and trees and lamppost.Emulator root Can be true according to the scene of the position of the barrier of sensor collection, shape, size, and the generation such as the gradient of road, topography variation Road environment when real reduction vehicle travels.So movement of the dynamic model such as pedestrian, vehicle, and the change etc. of road, Can truly it be showed in analogue system., can be by combining each road when being related to a plurality of road Scene generates road environment.And the scene that the dynamic models such as Pedestrians and vehicles are formed can be added to according to test needs In different road environments.Rich and varied test scene can be so simulated, just can be real only by the superposition of scene The road environment of analogue simulation road environment at that time and each class testing, and then verify planning control algorithm.
The present invention can establish model and scene by VR technologies.Also, the combination of scene and model can also pass through VR technologies are realized.There are detailed modeling method, and the method using model buildings scene in existing VR technologies, because This, no longer repeats the generating process of model and scene in the present invention.The emphasis of the present invention is that proposition can be by dynamic field Scape and static scene are mutually fused in a virtual environment, to form road environment more true to nature.The process of fusion can be with For the parameter of dynamic scene (such as coordinate, color, shape, size) is shown in analogue system into (by VR skill modelings), And be put into corresponding static scene, wherein, the parameter of dynamic scene is real-time change over time.Due to, dynamically with it is quiet The parameter of state scene is all the sensing data collected on the spot, therefore the reduction degree of the scene of combination producing is high.So, emulating In system, by the combination of dynamic scene and static scene, any road environment can be simulated, reaches and truly goes back original path ring Border, carry out the purpose of access control planning algorithm.
The present embodiment is it is further preferred that emulation module 102 can be emulator.
In the present embodiment, the sensing data by sensor collection is passed in emulator.Road is generated by emulator The scene of environment and vehicle drive carrys out verification algorithm.The scene of road environment can be by the built-up pattern in emulator, and root Model is put into map composition scene according to absolute location coordinates, then different scene maps is combined and restores overall road Road environment.Pretreatment module 101 can be process chip.The information of road surface of scene map then can be by process chip to sensing Data modeling generates.Because the emulation playback of the present invention is to build Scene realization, thus energy by the sensing data of collection The true scene restored at that time.Different scene maps are for example, can include dynamic scene and static scene.Dynamic scene and The means of the combination of static scene are for example, dynamic scene is merged with static scene according to coordinate information.
Fig. 3 illustrates the schematic block diagram of the emulation module 102 according to the embodiment of the present application.
As shown in figure 3, emulation module 102 can include scenario building unit 301, for receiving model and building scene.
In the present embodiment, scenario building unit 301 can be used for receiving dynamic model and static models, and to dynamic Model and static models are combined, and to generate dynamic scene and static scene, and dynamic scene and static scene are carried out Combination, to realize the reduction of road environment.
The major function of scenario building unit 301 is as follows:
Receive dynamic model and static models.By dynamic model according to test needs or the position relationship of original scene (absolute location information) is combined, and is put into same global coordinate system, generates dynamic scene.Static models are needed according to test Want or the position relationship (absolute location information) of original scene is combined, be put into same global coordinate system, generation is static Scene.
As shown in figure 3, emulation module 102 can also include coordinate transformation unit 303, for carrying out sensing data from complete Office coordinate to local coordinate conversion, to generate relative position information.
In the present embodiment, because the main purpose of dynamic simulation is to reduce certain vehicle A driving procedure.Therefore, coordinate Converter unit 303 is mainly used in dynamic scene being transformed under vehicle axis system from global coordinate system, to carry out emulation playback.Cause This, vehicle axis system may be considered the coordinate system on the basis of vehicle A.In addition, static scene does not change over time, therefore nothing Real-time coordinates conversion need to be carried out to static scene.Vehicle axis system is the vehicle A transverse direction and vertical using vehicle A as coordinate origin To for reference axis.Global coordinate system is to describe the positional information of earth surface object so that earth point is coordinate origin.
As shown in figure 3, emulation module 102 can also include dynamic simulation unit 302, for receive planning control information and The sensing data after conversion from world coordinates to local coordinate, and emulated for the dynamic data in sensing data, come The traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene.Planning control information can include artificial change The means of the planning control of vehicle, the planning control of Automatic Optimal vehicle., can be according to the time of each frame come automatic during emulation Dynamic data is extracted, generates dynamic scene.
The present embodiment it is further preferred that a kind of dynamic scene information recording and emulation playback reproducer, in addition to:Storage Module, stored for receiving dynamic scene and/or static scene, and to dynamic scene and/or static scene, to be provided with Combine scenes are used directly to during post-simulation, to realize the reduction of road environment.
So, dynamic scene and static scene are overlapped mutually into combination using VR technologies by scenario building unit 301 to exist Together, the fixation test scene strong with regard to reduction degree high stability can be generated, wherein, the fixation test scene is mainly used in simulating road Road environment, test the failure problems of vehicle.Memory module is used to store dynamic scene and static scene.Dynamic scene and static field Scape is used to combine fixed test scene.On the one hand, dynamic scene and static scene are stored, is easy to neatly combine dynamic Scene and static scene, to simulate miscellaneous road environment, generate abundant fixation test scene.On the other hand, to dynamic State scene and static scene are stored, convenient to generate reduction degree and the high fixation test scene of stability whenever and wherever possible.
The dynamic scene of the present invention and the structure of static scene are independent of each other independently of each other.Dynamic scene establish independent of Static data.It is more convenient for being individually created dynamic model and static mould if individually dynamic data and static data is handled Type, the scene of combinatorial path environment are more convenient, and also easily facilitate preservation.The model that the present invention can generate dynamic static data, And the scene that model is combined into preserves, dynamic data can be both preserved, static data can also be preserved, it is imitative after convenient True road environment.
Dynamic simulation unit 302, it is mainly used in the driving procedure of dynamic simulation vehicle.For example, update and show in real time Go out vehicle and the state with peripheral obstacle, the traveling process of vehicle is gone out come truly analogue simulation.Dynamic simulation unit 302, The relative position information that coordinate transformation unit 303 converts is received, the travel route of vehicle is constantly updated according to relative position information With the mobile route of barrier.Also, dynamic simulation unit 302 can also some be real-time by shock of vehicle and barrier etc. Physical effect display renewal is observed on screen for tester.Dynamic simulation unit can also according to planning control information come Operational order is parsed, and then controls the driving mode and travel route of vehicle.In addition, dynamic simulation unit can also possess to car Impact, object expansion, wind speed effect, light effects, the display and processing of the environmental factor such as topography variation.To above-mentioned environment because The analyzing and processing of element, may rely on the measured data of vehicle, system is by these sensing datas come analysis environments factor to driving The influence sailed, and be shown on emulation screen.The process of display and emulation may be referred to the thing in VR systems of the prior art The process that body interacts.
Therefore, tester can be observed the traveling process of vehicle in analogue system, and test-schedule control is calculated After method changes, vehicle whether can avoiding obstacles, and then improved planning control algorithm is verified.
The present embodiment is it is further preferred that planning control module includes:Unmanned Systems, for being believed according to planning control Breath and relative position information carry out planning control vehicle, to verify improved planning control algorithm, wherein, planning control information includes The means of the means of the artificial planning control for changing vehicle and the planning control of Automatic Optimal vehicle.
In the present embodiment, planning control module 103 can be Unmanned Systems.Planning control module 103 can be controlled The mobile route and drive manner of vehicle processed.The planning control algorithm of manufacturer's input is provided with planning control module 103.Tester Member corrects the mobile route and drive manner of vehicle by amending plans control algolithm.Planning control module 103 and planning are controlled Algorithm processed may be referred to the Unmanned Systems of prior art.In the present invention, planning control module 103 is not by planning control Information is exported to vehicle, but is exported to emulation module 102.So emulation module 102 could read the operating procedure of vehicle Emulated.The parameters such as the relative position information by calculating are inputted planning control module 103 to simulate car by emulation module 102 When intelligence system meets with certain road environment, process that vehicle is made a response automatically.Emulation module 102 can be from planning control The operational order of vehicle is parsed in information, and matches specific sensing data, effect caused by specific sensing data is shown Show in analogue system.Planning control information is used to realize the planning control to vehicle in emulation module 102.Emulation module 102 The sensing data of renewal is constantly transmitted to the change that planning control module 103 carrys out simulated roadway environment.Planning control module 103 is real When receive description road environment change sensing data and processing, according to planning control algorithm come do not stop correct vehicle traveling road Line and drive manner.Planning control module 103 can be just analyzed road environment automatically according to planning control algorithm, and is selected Suitable drive manner is selected, and traffic route is corrected automatically.
Compared with prior art, the recording of dynamic scene information according to embodiments of the present invention and emulation back method, dress Put, electronic equipment and computer-readable recording medium, give a kind of sensing data using real vehicle collection to rebuild true field The method of scape, to realize the regression test to planning control algorithm.The present invention can play the reality collected in drive test as far as possible The maximum effect of border data, greatly reduces the testing cost of intelligent driving, and realizes the reconstruction of real scene.The present invention is logical Cross and be transformed into the sensing data of real vehicle collection in global coordinate system from local coordinate system, and to sensing in emulation module 102 Data carry out the conversion from global coordinate system to local coordinate system, to realize the reconstruction of real scene, wherein, local coordinate system can Think vehicle axis system.In addition, by the way that the sensing data gathered under different condition is merged in emulation module 102, this Invention can also reconstruct various complicated scenes to verify planning control algorithm.
The present invention without by real steering vectors come innovatory algorithm, and using the magnanimity real vehicle data of passing collection with regard to that can reach Real-time scene rebuilds the purpose with emulation, can save time human cost, and and can generates complicated and abundant test scene, with Planning control algorithm is verified, so as to greatly improve the robustness of algorithm.
Illustrative methods
The recording and emulation of dynamic scene information according to the embodiment of the present application are described below with reference to Fig. 4, Fig. 5 and Fig. 6 The method of playback.
The invention provides a kind of recording of dynamic scene information and emulation back method, including:
Sensing data is collected, the conversion from local coordinate to world coordinates is carried out to sensing data, to generate absolute position Information;
Scene is established according to absolute location information, to realize the reduction of road environment, and sensing data is carried out from World coordinates to local coordinate conversion, to generate relative position information, and according to relative position information come analogue simulation car Traveling process, to realize the reduction of user's Driving Scene;
According to planning control information and relative position information come planning control vehicle, to verify that improved planning control is calculated Method, wherein, planning control information includes the artificial means of planning control and the planning control of Automatic Optimal vehicle for changing vehicle Means.
In the present embodiment, sensing data is collected, the conversion from local coordinate to world coordinates is carried out to sensing data, with Absolute location information is generated, above-mentioned steps may be referred to the worked of the pretreatment module 101 in the exemplary means of the present invention Journey (course of work of pretreatment module 101 can include the direct transform of coordinate described in exemplary means, sensing data it is whole Volume modeling etc.), the present invention is no longer repeated this.
Scene is established according to absolute location information, to realize the reduction of road environment, and sensing data is carried out from World coordinates to local coordinate conversion, to generate relative position information, and according to relative position information come analogue simulation car Traveling process, to realize the reduction of user's Driving Scene, above-mentioned steps may be referred in the exemplary means of the present invention (course of work of emulation module 102 can be including the coordinate described in exemplary means just for the course of work of emulation module 102 Conversion, structure scene, combine scenes generation road environment, the traveling process of analogue simulation vehicle etc.), the present invention no longer enters to this Row repeats.
According to planning control information and relative position information come planning control vehicle, to verify that improved planning control is calculated Method, wherein, planning control information includes the artificial means of planning control and the planning control of Automatic Optimal vehicle for changing vehicle Means, above-mentioned steps may be referred to the present invention exemplary means in planning control module 103 the course of work (planning control The course of work of molding block 103 can include the control mode of planning control module 103 described in exemplary means, and planning Control module 103 and the interaction of emulation module 102 etc.), the present invention is no longer repeated this.
In the present invention, relative position information can include the real-time vehicle body positional information and barrier under first person Hinder object location information.Absolute location information can include the real-time vehicle body positional information under third person and barrier position Confidence ceases.
The present embodiment is it is further preferred that the method for the present invention can also include identifying, moving static number for dynamic static data According to modeling, and static models and dynamic model are generated respectively, for example, specifically, may include steps of:
Dynamic Data Processing and static data processing are carried out to the sensing data under global coordinate system, to generate dynamic model And static models.Dynamic Data Processing and static data processing can be carried out respectively to sensing data, such as sensing data is entered Row identification, and cover timestamp.In the present invention, timestamp is used for representing the acquisition time of sensing data, and can be convenient Analogue system temporally exports dynamic data.Sensing data is identified can be by the identification of existing dynamic static data Method, for example, can by adding dynamic labels and static labels respectively to the dynamic data in sensing data and static data, To identify dynamic data and static data.
Above-mentioned steps can also refer to the dynamic static data of the identification of pretreatment module 101 in the exemplary means of the present invention, And the process of generation static models and dynamic model, the present invention are no longer repeated this.
The present embodiment is it is further preferred that the method for the present invention can also include building scene, life by dynamic static data Into static scene and dynamic scene, combine dynamic static scene and generate road environment, the traveling process of analogue simulation vehicle, example Such as, specifically, may include steps of:
Dynamic model and static models are combined according to absolute location information, to generate dynamic scene and static field Scape.Dynamic scene and static scene are combined, to realize the reduction of road environment.Dynamic scene to generation and/or quiet State scene is stored, for be used directly to combine scenes during post-simulation, to realize the reduction of road environment.To dynamic scene Mode is combined with static scene, for example, dynamic scene is merged with static scene according to coordinate information, to realize The reduction of road environment.
Conversion from world coordinates to local coordinate is carried out to sensing data, to generate relative position information, and according to Relative position information carrys out the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene.
The emulation module 102 that above-mentioned steps can also be referred in the exemplary means of the present invention generates static scene and dynamic The process of scene, and the process of the dynamic static scene generation road environment of combination, the traveling process of analogue simulation vehicle, the present invention This is no longer repeated.
The present embodiment is it is further preferred that carry out the conversion from local coordinate to world coordinates or global seat to sensing data Marking the conversion of local coordinate includes:It is online to change sensing data in real time and/or change sensing data offline.
In the present embodiment, the scheme of the online sensing data of conversion in real time is for example, during real vehicle gathered data, in real time to coordinate Direct transform and inverse transformation are carried out, calculates the position of vehicle and barrier, and real-time update simulation result and is shown.Offline conversion Sensing data from cloud platform for example, download mass data to local, these data are pre-processed in cloud platform, such as are sat Mark direct transform.Emulator reads local data, carries out inverse transformation to coordinate, calculates the position of vehicle and barrier, and in real time Renewal simulation result is simultaneously shown.
The present embodiment is it is further preferred that the traveling process of analogue simulation vehicle only includes the emulation to dynamic data Journey, to save calculation resources, for example, receiving the sensing number after planning control information and conversion from world coordinates to local coordinate According to, and emulated for the dynamic data in sensing data, carry out the traveling process of analogue simulation vehicle, to realize that user drives The reduction of scene.Above-mentioned steps can also refer to the course of work of the dynamic simulation unit 302 in the exemplary means of the present invention, The present invention is no longer repeated this.
Fig. 6 illustrates the step schematic diagram of recording and the emulation playback according to the dynamic scene information of the embodiment of the present application.
As shown in fig. 6, the present embodiment is further preferably, the method for recording of dynamic scene information can include:Step S100:Collect sensing data.Step S200:Sensing data is pre-processed.Step S300:Sensing data is carried out from vehicle Coordinate transform of the coordinate system to global coordinate system.Step S400:Static scene is built according to pretreated sensing data.Step S500:The coordinate for carrying out transforming to vehicle axis system to dynamic scene according to the position of vehicle in emulation from global coordinate system becomes Change.Step S600:Planning control information is received, wherein, planning control information includes the artificial planning control for changing vehicle, automatic Optimize the method for the planning control of vehicle.Step S700:The dynamic scene at corresponding time point is extracted according to temporal information, by sitting After mark conversion, the dynamic simulation playback of vehicle operation is carried out.
In the present embodiment, sensing data can include data type label and timestamp, wherein, timestamp is used to represent The acquisition time of sensing data.Data type label includes dynamic labels and static labels.
In the present embodiment, vehicle axis system is using vehicle as coordinate origin, and the horizontal and vertical of vehicle is reference axis. Global coordinate system describes the positional information of earth surface object so that earth point is coordinate origin.
Pretreatment can include the direct transform of coordinate and the modeling of sensing data.Modeling includes the structure of dynamic model and quiet The structure of states model.The order of modeling be it is random, can also synchronization process move and static data and model.Build static models and Structure dynamic model can individually be handled and can also handled together, be closed because having no dependence between dynamic data and static data System.
Build static scene and structure dynamic scene can with synchronization process, can also serial process, be not subject to processing order Influence.Structure static scene and structure dynamic scene can individually be handled and can also handled together, because dynamic data and static state Dependence is had no between data.
Dynamic data, being capable of accurate description for example, dynamic sensing data and the dynamic model and dynamic scene of generation The movement locus that the dynamic object changes over time.Sensing data can be preserved with time sequencing, so that being capable of basis in emulation Time extracts dynamic data.Static data is for example, static sensing data and the static models and static scene of generation, no It can change over time.
Fig. 4 describes the critical workflow that sensing data of the present invention is collected and pre-processed.As shown in figure 4, the recording and emulation Method can include:Step S1:Sensor gathers sensing data.Step S2:Collect the sensing data collected.Step S3 ':Note Record the sensing data collected.Step S3:Sensing data is exported and carries out planning control.Step S4:Sensing data is carried out pre- Processing, and coordinate transform is carried out to sensing data, vehicle axis system is transformed into global coordinate system.Sensor can include swashing Optical radar, vision sensor and ultrasonic sensor etc..
In the present embodiment, within the planning control cycle, sensing data is sent to planning control module, while data are received Collection module records sensing data in real time.In step s 4, data collection module by sensing data from vehicle body local coordinate system transfer Change in global coordinate system.The step can be changed in real time online, can also be turned offline according to the vehicle body positional information of record Change to save computing resource.
Fig. 5 illustrates the critical workflow of the emulation playback of the present invention.As shown in figure 5, the recording and emulation mode can wrap Include:Step S5:Collect dynamic playback data.Step S6:Dynamic playback data in data type is respectively transmitted to emulator.Number It can include laser radar data, visual sensing data and supersonic sensing data according to type.Emulator carries out dynamic scene Coordinate transform, for example, carry out coordinate transform to dynamic scene according to the position of vehicle in emulation (transforms to car from global coordinate system Coordinate system).Step S5 ':Collect static playback of data.Step S6 ':By static playback of data by data type be respectively transmitted to Emulator.Emulator builds static scene and dynamic scene according to pretreated sensing data.By dynamic scene and static field Scape is merged according to coordinate information, comes assembled static scene and dynamic scene.Data type can include laser radar data, Visual sensing data and supersonic sensing data.Step S7:Emulator receives the planning control information of Unmanned Systems, and imitative True device carries out dynamic simulation playback.Dynamic simulation plays back for example, emulator reading planning control information, is extracted according to temporal information The dynamic scene at corresponding time point, after coordinate transform, carry out dynamic simulation playback.
In the present embodiment, Unmanned Systems can control the driving procedure of vehicle in emulator, such as unmanned System exports planning control information and carries out planning control to vehicle.Emulator sends the sensing datas such as vehicle body position to nobody and driven Sail system.Unmanned Systems carry out planning control according to sensing data to vehicle.Planning control information is for example, artificially change car The means such as planning control, the planning control of Automatic Optimal vehicle.
Compared with prior art, the recording of dynamic scene information according to embodiments of the present invention and emulation back method, dress Put, electronic equipment and computer-readable recording medium, give a kind of sensing data using real vehicle collection to rebuild true field The method of scape, to reach the regression test to planning control algorithm minor modifications.The present invention can be played in drive test as far as possible The maximum effect of the real data of collection, greatly reduces the testing cost of intelligent driving, and realizes the reconstruction of real scene. The present invention is by the way that real vehicle sensing data is transformed into global coordinate system from vehicle body local coordinate system, and according to car in emulator Positional information reconvert carriage return body local coordinate system, to realize the reconstruction of real scene.In addition, by will be adopted under different condition Fusion of the sensing data of collection in emulator, the present invention can also reconstruct various complicated scenes to verify that planning control is calculated Method.
The present invention without by real steering vectors come innovatory algorithm, and using the magnanimity real vehicle data of passing collection with regard to that can reach Real-time scene rebuilds the purpose with emulation, can save time human cost, and and can greatly improves the level of coverage of checking scene, The iteration of planning control algorithm is verified, the robustness of algorithm can be greatly improved.
Example electronic device
The electronic equipment of the embodiment of the present application.The electronic equipment can include the A devices for being used for Data Collection and pretreatment With for emulate playback B devices in any one or both.The electronic equipment can also be set including the unit independent with them Standby C (such as Unmanned Systems), the stand-alone device can be communicated with A devices and B devices, to be connect from A devices and B devices Input signal is received, and input signal is handled and fed back.
Electronic equipment can include one or more sensors and processor.
Processor can be CPU (CPU) or have data-handling capacity and/or instruction execution capability The processing unit of other forms, and can be with the other assemblies in control electronics to perform desired function.
Sensor can be the equipment such as binocular camera, laser radar, ultrasonic sensor.
Electronic equipment can also include one or more memories.Memory can include one or more computer programs Product, the computer program product can include various forms of computer-readable recording mediums, such as volatile memory And/or nonvolatile memory.The volatile memory can for example include random access memory (RAM) and/or high speed Buffer storage (cache) etc..The nonvolatile memory can for example include read-only storage (ROM), hard disk, flash memory Deng.One or more computer program instructions can be stored on the computer-readable recording medium, processor can be run Described program instructs, to realize the recording of the dynamic scene information of each embodiment of the application described above and emulation mode And/or other desired functions.Such as input signal, signal can also be stored in the computer-readable recording medium The various contents such as component.
In one example, electronic equipment can also include:Input unit and output device, these components pass through total linear system The interconnection of the bindiny mechanism of system and/or other forms.
For example, when the electronic equipment is A devices or B devices, the input unit can include sensor.Input unit will Sensing data is transferred to A devices or B devices.When the electronic equipment is stand-alone device C (such as Unmanned Systems), the input Device can be communication network connector, for from the sensing data after A devices or B device reception processings.
In addition, the input equipment can also include such as keyboard, mouse etc..
The output device can export various information, including vehicle body position, control command etc. to outside.The output equipment can With including such as display, loudspeaker, printer and communication network and its remote output devices connected etc..
Illustrative computer program product and computer-readable recording medium
In addition to the above method and equipment, embodiments herein can also be computer program product, and it includes meter Calculation machine programmed instruction, the computer program instructions cause that described computing device this specification is above-mentioned when being run by processor In the recording and emulation mode of the dynamic scene information according to the various embodiments of the application described in " illustrative methods " part The step of.
The computer program product can be used to hold with any combination of one or more programming languages to write The program code of row the embodiment of the present application operation, described program design language include object oriented program language, such as Java, C++ etc., in addition to conventional procedural programming language, such as " C " language or similar programming language.Journey Sequence code can perform fully on the user computing device, partly perform on a user device, independent as one soft Part bag performs, part performs or completely in remote computing device on a remote computing on the user computing device for part Or performed on server.
In addition, embodiments herein can also be computer-readable recording medium, it is stored thereon with computer program and refers to Order, the computer program instructions cause above-mentioned " the exemplary side of described computing device this specification when being run by processor Step in the recording and emulation mode of the dynamic scene information according to the various embodiments of the application described in method " part.
The computer-readable recording medium can use any combination of one or more computer-readable recording mediums.Computer-readable recording medium can To be readable signal medium or readable storage medium storing program for executing.Readable storage medium storing program for executing can for example include but is not limited to electricity, magnetic, light, electricity Magnetic, the system of infrared ray or semiconductor, device or device, or any combination above.Readable storage medium storing program for executing is more specifically Example (non exhaustive list) includes:Electrical connection, portable disc with one or more wires, hard disk, random access memory Device (RAM), read-only storage (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc Read-only storage (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
The general principle of the application is described above in association with specific embodiment, however, it is desirable to, it is noted that in this application The advantages of referring to, advantage, effect etc. are only exemplary rather than limiting, it is impossible to which it is the application to think these advantages, advantage, effect etc. Each embodiment is prerequisite.In addition, detail disclosed above is merely to the effect of example and the work readily appreciated With, and it is unrestricted, above-mentioned details is not intended to limit the application as that must be realized using above-mentioned concrete details.
The device that is related in the application, device, equipment, the block diagram of system only illustratively the example of property and are not intended to It is required that or hint must be attached in the way of square frame illustrates, arrange, configure.As it would be recognized by those skilled in the art that , it can connect, arrange by any-mode, configuring these devices, device, equipment, system.Such as " comprising ", "comprising", " tool Have " etc. word be open vocabulary, refer to " including but is not limited to ", and can be with its used interchangeably.Vocabulary used herein above "or" and " and " refer to vocabulary "and/or", and can be with its used interchangeably, unless it is not such that context, which is explicitly indicated,.Here made Vocabulary " such as " refers to phrase " such as, but not limited to ", and can be with its used interchangeably.
It may also be noted that in device, apparatus and method in the application, each part or each step are to decompose And/or reconfigure.These decompose and/or reconfigured the equivalents that should be regarded as the application.
The above description of disclosed aspect is provided so that any person skilled in the art can make or use this Application.Various modifications in terms of these are readily apparent to those skilled in the art, and defined herein General Principle can apply to other aspect without departing from scope of the present application.Therefore, the application is not intended to be limited to Aspect shown in this, but according to the widest range consistent with the feature of principle disclosed herein and novelty.
In order to which purpose of illustration and description has been presented for above description.In addition, this description is not intended to the reality of the application Apply example and be restricted to form disclosed herein.Although already discussed above multiple exemplary aspects and embodiment, this area skill Art personnel will be recognized that its some modifications, modification, change, addition and sub-portfolio.

Claims (10)

1. recording and the emulation back method of a kind of dynamic scene information, including:
Sensing data is collected, the conversion from local coordinate to world coordinates is carried out to sensing data, to generate absolute location information, Wherein, absolute location information is positional information of the target under global coordinate system;
Scene is established according to absolute location information, to realize the reduction of road environment, and sensing data is carried out from the overall situation Coordinate to local coordinate conversion, to generate relative position information, and according to relative position information come analogue simulation vehicle Traveling process, to realize the reduction of user's Driving Scene, wherein, relative position information is position of the target under local coordinate system Information;
According to planning control information and relative position information come planning control vehicle, to verify improved planning control algorithm, its In, planning control information includes the planning control for the means and/or Automatic Optimal vehicle for artificially changing the planning control of vehicle Means.
2. the method for claim 1, wherein establishing scene includes:
Dynamic Data Processing and static data processing are carried out to the sensing data under global coordinate system, to generate dynamic model and quiet States model;
Dynamic model and static models are combined, to generate dynamic scene and static scene;
Dynamic scene and static scene are combined, to realize the reduction of road environment.
3. method as claimed in claim 2, wherein, establishing scene also includes:
The dynamic scene and/or static scene of generation are stored, for be used directly to combine scenes during post-simulation, with reality The reduction of existing road environment.
4. the method for claim 1, wherein the traveling process of analogue simulation vehicle includes:
The sensing data after planning control information and conversion from world coordinates to local coordinate is received, and in sensing data Dynamic data emulated, carry out the traveling process of analogue simulation vehicle, to realize the reduction of user's Driving Scene.
5. recording and the emulation playback reproducer of a kind of dynamic scene information, including:
Pretreatment module, for carrying out the conversion from local coordinate to world coordinates to sensing data, to generate absolute position letter Breath;
Emulation module, for establishing scene according to absolute location information, to realize the reduction of road environment, and to sensing number According to conversion from world coordinates to local coordinate is carried out, to generate relative position information, and imitated according to relative position information The traveling process of true simulating vehicle, to realize the reduction of user's Driving Scene;
Planning control module, for come planning control vehicle, being improved according to planning control information and relative position information with verifying Planning control algorithm, wherein, planning control information includes the means and/or Automatic Optimal of the artificial planning control for changing vehicle The means of the planning control of vehicle.
6. device as claimed in claim 5, wherein, the pretreatment module includes:
Data processing module, for carrying out dynamic number to the dynamic data in the sensing data under global coordinate system and static data According to processing and static data processing, to generate dynamic model and static models.
7. device as claimed in claim 6, wherein, the emulation module includes:
Scenario building unit, it is combined for receiving dynamic model and static models, and to dynamic model and static models, To generate dynamic scene and static scene, and dynamic scene and static scene are combined, to realize going back for road environment It is former.
8. device as claimed in claim 5, wherein, the emulation module includes:
Dynamic simulation unit, for receiving the sensing number after planning control information and conversion from world coordinates to local coordinate According to, and emulated for the dynamic data in sensing data, carry out the traveling process of analogue simulation vehicle, to realize that user drives The reduction of scene.
9. a kind of electronic equipment, including:
Processor;And
Memory, computer program instructions are stored with the memory, the computer program instructions are by the processing Device causes method of the computing device as any one of claim 1-4 when running.
10. a kind of computer-readable recording medium, computer program instructions are stored thereon with, the computer program instructions are in quilt Processor causes method of the computing device as any one of claim 1-4 when running.
CN201710930672.6A 2017-10-09 2017-10-09 Dynamic scene information recording and simulation playback method, device, equipment and medium Active CN107678306B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710930672.6A CN107678306B (en) 2017-10-09 2017-10-09 Dynamic scene information recording and simulation playback method, device, equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710930672.6A CN107678306B (en) 2017-10-09 2017-10-09 Dynamic scene information recording and simulation playback method, device, equipment and medium

Publications (2)

Publication Number Publication Date
CN107678306A true CN107678306A (en) 2018-02-09
CN107678306B CN107678306B (en) 2021-04-16

Family

ID=61139355

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710930672.6A Active CN107678306B (en) 2017-10-09 2017-10-09 Dynamic scene information recording and simulation playback method, device, equipment and medium

Country Status (1)

Country Link
CN (1) CN107678306B (en)

Cited By (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109146898A (en) * 2018-09-07 2019-01-04 百度在线网络技术(北京)有限公司 A kind of simulation number Enhancement Method, device and terminal
CN109598066A (en) * 2018-12-05 2019-04-09 百度在线网络技术(北京)有限公司 Effect evaluation method, device, equipment and the storage medium of prediction module
CN109606384A (en) * 2018-12-29 2019-04-12 百度在线网络技术(北京)有限公司 Control method for vehicle, device, equipment and storage medium
CN109885929A (en) * 2019-02-19 2019-06-14 百度在线网络技术(北京)有限公司 Automatic Pilot decision rule data reproducing method and device
CN109933856A (en) * 2019-02-19 2019-06-25 百度在线网络技术(北京)有限公司 Automatic Pilot decision rule data reproducing method and device
CN109960857A (en) * 2019-02-28 2019-07-02 东软睿驰汽车技术(沈阳)有限公司 A kind of emulation mode of laser radar, device and emulation platform
EP3618008A1 (en) * 2018-08-30 2020-03-04 Baidu Online Network Technology (Beijing) Co., Ltd. Method for generating simulated point cloud data, device, and storage medium
CN111123731A (en) * 2018-10-31 2020-05-08 百度在线网络技术(北京)有限公司 Method, device, storage medium and terminal equipment for simulating automatic driving vehicle
CN111123732A (en) * 2018-10-31 2020-05-08 百度在线网络技术(北京)有限公司 Method, device, storage medium and terminal equipment for simulating automatic driving vehicle
CN111143377A (en) * 2018-11-05 2020-05-12 百度在线网络技术(北京)有限公司 Automatic driving simulation data collection method, device and system
CN111159893A (en) * 2019-12-30 2020-05-15 北京理工大学 Fault recurrence and prediction system based on digital twinning
CN111199087A (en) * 2018-10-31 2020-05-26 百度在线网络技术(北京)有限公司 Scene recognition method and device
CN111199088A (en) * 2018-10-31 2020-05-26 百度在线网络技术(北京)有限公司 Method and device for reproducing scene data
CN111367194A (en) * 2018-12-25 2020-07-03 北京欣奕华科技有限公司 Visual algorithm verification method and device
CN111580493A (en) * 2020-04-14 2020-08-25 吉利汽车研究院(宁波)有限公司 Automatic driving simulation method, system and medium
CN111579251A (en) * 2020-04-16 2020-08-25 国汽(北京)智能网联汽车研究院有限公司 Method, device and equipment for determining vehicle test scene and storage medium
CN111665738A (en) * 2019-03-05 2020-09-15 阿里巴巴集团控股有限公司 In-loop simulation system and information processing method and device thereof
CN112567374A (en) * 2020-10-21 2021-03-26 华为技术有限公司 Simulated traffic scene file generation method and device
US10984588B2 (en) 2018-09-07 2021-04-20 Baidu Online Network Technology (Beijing) Co., Ltd Obstacle distribution simulation method and device based on multiple models, and storage medium
CN112881052A (en) * 2021-01-14 2021-06-01 深圳市杉川机器人有限公司 Method and device for constructing working scene of mobile robot
US11047673B2 (en) 2018-09-11 2021-06-29 Baidu Online Network Technology (Beijing) Co., Ltd Method, device, apparatus and storage medium for detecting a height of an obstacle
US11113546B2 (en) 2018-09-04 2021-09-07 Baidu Online Network Technology (Beijing) Co., Ltd. Lane line processing method and device
US11126875B2 (en) 2018-09-13 2021-09-21 Baidu Online Network Technology (Beijing) Co., Ltd. Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium
CN113626317A (en) * 2021-07-28 2021-11-09 的卢技术有限公司 Automatic driving software debugging system, method, medium and equipment
CN113721475A (en) * 2020-05-26 2021-11-30 广州汽车集团股份有限公司 Data exchange method, data exchange control device and data exchange system
US11205289B2 (en) 2018-09-07 2021-12-21 Baidu Online Network Technology (Beijing) Co., Ltd. Method, device and terminal for data augmentation
CN113947893A (en) * 2021-09-03 2022-01-18 网络通信与安全紫金山实验室 Method and system for restoring driving scene of automatic driving vehicle
CN113949996A (en) * 2021-10-11 2022-01-18 重庆大学 Data processing method and system for acquisition and playback of vehicle-road cooperative test scene
US11307302B2 (en) 2018-09-07 2022-04-19 Baidu Online Network Technology (Beijing) Co., Ltd Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium
EP4109194A1 (en) * 2021-06-25 2022-12-28 Siemens Aktiengesellschaft Sensor data generation for controlling an autonomous vehicle
US11718318B2 (en) 2019-02-22 2023-08-08 Apollo Intelligent Driving (Beijing) Technology Co., Ltd. Method and apparatus for planning speed of autonomous vehicle, and storage medium
US11780463B2 (en) 2019-02-19 2023-10-10 Baidu Online Network Technology (Beijing) Co., Ltd. Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101077578A (en) * 2007-07-03 2007-11-28 北京控制工程研究所 Mobile Robot local paths planning method on the basis of binary environmental information
CN102426425A (en) * 2011-11-08 2012-04-25 重庆邮电大学 Automobile ABS (Antilock Brake System) virtual reality simulation system
CN103298728A (en) * 2011-07-05 2013-09-11 天宝导航有限公司 Crane maneuvering assistance
CN104596770A (en) * 2015-01-20 2015-05-06 天津大学 Comprehensive performance testing system for vehicle power assembly
CN104655007A (en) * 2013-11-22 2015-05-27 中国科学院深圳先进技术研究院 Method and system for creating world coordinates of environment scene
CN104900111A (en) * 2015-06-26 2015-09-09 苏州大学张家港工业技术研究院 High fidelity simulation driving system and method
CN106021750A (en) * 2016-05-25 2016-10-12 北京经纬恒润科技有限公司 Generation method and generation system for simulation scene and reconstruction unit thereof
CN106644503A (en) * 2016-10-24 2017-05-10 中国科学院合肥物质科学研究院 Intelligent vehicle planning capacity testing platform
CN106940942A (en) * 2017-05-12 2017-07-11 南京全控航空科技有限公司 The driving training simulator and method of self-adaptative adjustment Training scene
US20170248963A1 (en) * 2015-11-04 2017-08-31 Zoox, Inc. Adaptive mapping to navigate autonomous vehicles responsive to physical environment changes

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101077578A (en) * 2007-07-03 2007-11-28 北京控制工程研究所 Mobile Robot local paths planning method on the basis of binary environmental information
CN103298728A (en) * 2011-07-05 2013-09-11 天宝导航有限公司 Crane maneuvering assistance
CN102426425A (en) * 2011-11-08 2012-04-25 重庆邮电大学 Automobile ABS (Antilock Brake System) virtual reality simulation system
CN104655007A (en) * 2013-11-22 2015-05-27 中国科学院深圳先进技术研究院 Method and system for creating world coordinates of environment scene
CN104596770A (en) * 2015-01-20 2015-05-06 天津大学 Comprehensive performance testing system for vehicle power assembly
CN104900111A (en) * 2015-06-26 2015-09-09 苏州大学张家港工业技术研究院 High fidelity simulation driving system and method
US20170248963A1 (en) * 2015-11-04 2017-08-31 Zoox, Inc. Adaptive mapping to navigate autonomous vehicles responsive to physical environment changes
CN106021750A (en) * 2016-05-25 2016-10-12 北京经纬恒润科技有限公司 Generation method and generation system for simulation scene and reconstruction unit thereof
CN106644503A (en) * 2016-10-24 2017-05-10 中国科学院合肥物质科学研究院 Intelligent vehicle planning capacity testing platform
CN106940942A (en) * 2017-05-12 2017-07-11 南京全控航空科技有限公司 The driving training simulator and method of self-adaptative adjustment Training scene

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
叶焕倬: "《三维运动目标的数据组织与管理》", 30 December 2007, 湖北科学技术出版社 *
李晓芸: "考虑避撞的多智能车平台动态路径跟踪控制研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *
李白云: "面向机器人自主导航的仿真关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 *
荆旭: "基于虚拟现实技术的汽车虚拟驾驶系统的研究与开发", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *

Cited By (44)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3618008A1 (en) * 2018-08-30 2020-03-04 Baidu Online Network Technology (Beijing) Co., Ltd. Method for generating simulated point cloud data, device, and storage medium
US11113830B2 (en) 2018-08-30 2021-09-07 Baidu Online Network Technology (Beijing) Co., Ltd. Method for generating simulated point cloud data, device, and storage medium
US11113546B2 (en) 2018-09-04 2021-09-07 Baidu Online Network Technology (Beijing) Co., Ltd. Lane line processing method and device
CN109146898A (en) * 2018-09-07 2019-01-04 百度在线网络技术(北京)有限公司 A kind of simulation number Enhancement Method, device and terminal
US11307302B2 (en) 2018-09-07 2022-04-19 Baidu Online Network Technology (Beijing) Co., Ltd Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium
US11276243B2 (en) 2018-09-07 2022-03-15 Baidu Online Network Technology (Beijing) Co., Ltd. Traffic simulation method, device and storage medium
US11205289B2 (en) 2018-09-07 2021-12-21 Baidu Online Network Technology (Beijing) Co., Ltd. Method, device and terminal for data augmentation
US10984588B2 (en) 2018-09-07 2021-04-20 Baidu Online Network Technology (Beijing) Co., Ltd Obstacle distribution simulation method and device based on multiple models, and storage medium
US11519715B2 (en) 2018-09-11 2022-12-06 Baidu Online Network Technology (Beijing) Co., Ltd. Method, device, apparatus and storage medium for detecting a height of an obstacle
US11047673B2 (en) 2018-09-11 2021-06-29 Baidu Online Network Technology (Beijing) Co., Ltd Method, device, apparatus and storage medium for detecting a height of an obstacle
US11126875B2 (en) 2018-09-13 2021-09-21 Baidu Online Network Technology (Beijing) Co., Ltd. Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium
CN111199087A (en) * 2018-10-31 2020-05-26 百度在线网络技术(北京)有限公司 Scene recognition method and device
CN111199088B (en) * 2018-10-31 2023-06-20 百度在线网络技术(北京)有限公司 Method and device for reproducing scene data
CN111199088A (en) * 2018-10-31 2020-05-26 百度在线网络技术(北京)有限公司 Method and device for reproducing scene data
CN111123731B (en) * 2018-10-31 2022-12-02 阿波罗智能技术(北京)有限公司 Method, device, storage medium and terminal equipment for simulating automatic driving vehicle
CN111123732A (en) * 2018-10-31 2020-05-08 百度在线网络技术(北京)有限公司 Method, device, storage medium and terminal equipment for simulating automatic driving vehicle
CN111123731A (en) * 2018-10-31 2020-05-08 百度在线网络技术(北京)有限公司 Method, device, storage medium and terminal equipment for simulating automatic driving vehicle
CN111143377B (en) * 2018-11-05 2024-06-11 阿波罗智能技术(北京)有限公司 Automatic driving simulation data collection method, device and system
CN111143377A (en) * 2018-11-05 2020-05-12 百度在线网络技术(北京)有限公司 Automatic driving simulation data collection method, device and system
CN109598066A (en) * 2018-12-05 2019-04-09 百度在线网络技术(北京)有限公司 Effect evaluation method, device, equipment and the storage medium of prediction module
CN109598066B (en) * 2018-12-05 2023-08-08 百度在线网络技术(北京)有限公司 Effect evaluation method, apparatus, device and storage medium for prediction module
CN111367194A (en) * 2018-12-25 2020-07-03 北京欣奕华科技有限公司 Visual algorithm verification method and device
CN109606384A (en) * 2018-12-29 2019-04-12 百度在线网络技术(北京)有限公司 Control method for vehicle, device, equipment and storage medium
US11215996B2 (en) 2018-12-29 2022-01-04 Apollo Intelligent Driving Technology (Beijing) Co., Ltd. Method and device for controlling vehicle, device, and storage medium
CN109885929A (en) * 2019-02-19 2019-06-14 百度在线网络技术(北京)有限公司 Automatic Pilot decision rule data reproducing method and device
CN109933856A (en) * 2019-02-19 2019-06-25 百度在线网络技术(北京)有限公司 Automatic Pilot decision rule data reproducing method and device
US11780463B2 (en) 2019-02-19 2023-10-10 Baidu Online Network Technology (Beijing) Co., Ltd. Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle
US11718318B2 (en) 2019-02-22 2023-08-08 Apollo Intelligent Driving (Beijing) Technology Co., Ltd. Method and apparatus for planning speed of autonomous vehicle, and storage medium
CN109960857A (en) * 2019-02-28 2019-07-02 东软睿驰汽车技术(沈阳)有限公司 A kind of emulation mode of laser radar, device and emulation platform
CN111665738A (en) * 2019-03-05 2020-09-15 阿里巴巴集团控股有限公司 In-loop simulation system and information processing method and device thereof
CN111159893A (en) * 2019-12-30 2020-05-15 北京理工大学 Fault recurrence and prediction system based on digital twinning
CN111580493A (en) * 2020-04-14 2020-08-25 吉利汽车研究院(宁波)有限公司 Automatic driving simulation method, system and medium
CN111580493B (en) * 2020-04-14 2022-08-30 吉利汽车研究院(宁波)有限公司 Automatic driving simulation method, system and medium
CN111579251A (en) * 2020-04-16 2020-08-25 国汽(北京)智能网联汽车研究院有限公司 Method, device and equipment for determining vehicle test scene and storage medium
CN113721475A (en) * 2020-05-26 2021-11-30 广州汽车集团股份有限公司 Data exchange method, data exchange control device and data exchange system
CN112567374A (en) * 2020-10-21 2021-03-26 华为技术有限公司 Simulated traffic scene file generation method and device
CN112881052A (en) * 2021-01-14 2021-06-01 深圳市杉川机器人有限公司 Method and device for constructing working scene of mobile robot
CN112881052B (en) * 2021-01-14 2024-02-20 深圳市杉川机器人有限公司 Method and device for constructing working scene of mobile robot
EP4109194A1 (en) * 2021-06-25 2022-12-28 Siemens Aktiengesellschaft Sensor data generation for controlling an autonomous vehicle
CN113626317B (en) * 2021-07-28 2023-09-05 的卢技术有限公司 Automatic driving software debugging system, method, medium and equipment
CN113626317A (en) * 2021-07-28 2021-11-09 的卢技术有限公司 Automatic driving software debugging system, method, medium and equipment
CN113947893A (en) * 2021-09-03 2022-01-18 网络通信与安全紫金山实验室 Method and system for restoring driving scene of automatic driving vehicle
CN113949996B (en) * 2021-10-11 2022-10-28 重庆大学 Data processing method and system for acquisition and playback of vehicle-road cooperative test scene
CN113949996A (en) * 2021-10-11 2022-01-18 重庆大学 Data processing method and system for acquisition and playback of vehicle-road cooperative test scene

Also Published As

Publication number Publication date
CN107678306B (en) 2021-04-16

Similar Documents

Publication Publication Date Title
CN107678306A (en) Dynamic scene information is recorded and emulation back method, device, equipment and medium
US10678244B2 (en) Data synthesis for autonomous control systems
US10482003B1 (en) Method and system for modifying a control unit of an autonomous car
CN109327695A (en) The test drives scene database system of virtual test Driving Scene for height reality
CN107844635B (en) System for realizing BIM information and traffic simulation information integration and integration method thereof
CN109285220A (en) A kind of generation method, device, equipment and the storage medium of three-dimensional scenic map
CN114879631A (en) Automatic driving test system and method based on digital twin cloud control platform
CN110103983A (en) System and method for the verifying of end-to-end autonomous vehicle
CN109324539A (en) The intelligent control platform and method of a kind of automatic Pilot closed test field
CN107145578A (en) Map constructing method, device, equipment and system
Gechter et al. Virtual intelligent vehicle urban simulator: Application to vehicle platoon evaluation
CN107144285A (en) Posture information determines method, device and movable equipment
Li et al. Advanced scenario generation for calibration and verification of autonomous vehicles
CN112286206A (en) Automatic driving simulation method, system, equipment, readable storage medium and platform
CN107085422A (en) A kind of tele-control system of the multi-functional Hexapod Robot based on Xtion equipment
CN101630146B (en) Simulation control system for remote tele-operation of lunar rover
CN108319249A (en) Unmanned algorithm synthesis evaluation system based on driving simulator and method
CN109726426A (en) A kind of Vehicular automatic driving virtual environment building method
WO2019065409A1 (en) Automatic driving simulator and map generation method for automatic driving simulator
CN114067062A (en) Method and system for simulating real driving scene, electronic equipment and storage medium
CN111752258A (en) Operation test of autonomous vehicle
CN113495495B (en) Construction method of track obstacle detection simulation platform based on illusion engine
CN110793548A (en) Navigation simulation test system based on virtual-real combination of GNSS receiver hardware in loop
CN109376664A (en) Machine learning training method, device, server and medium
Ramakrishna et al. Anti-carla: An adversarial testing framework for autonomous vehicles in carla

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant