EP3618013A1 - System zur erzeugung von fahrzeugsensordaten - Google Patents

System zur erzeugung von fahrzeugsensordaten Download PDF

Info

Publication number
EP3618013A1
EP3618013A1 EP18191020.9A EP18191020A EP3618013A1 EP 3618013 A1 EP3618013 A1 EP 3618013A1 EP 18191020 A EP18191020 A EP 18191020A EP 3618013 A1 EP3618013 A1 EP 3618013A1
Authority
EP
European Patent Office
Prior art keywords
sensor
data
model unit
simulation model
degraded
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.)
Withdrawn
Application number
EP18191020.9A
Other languages
English (en)
French (fr)
Inventor
Karthik Jagannath
Rainer Aue
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.)
Continental Automotive GmbH
Original Assignee
Continental Automotive GmbH
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 Continental Automotive GmbH filed Critical Continental Automotive GmbH
Priority to EP18191020.9A priority Critical patent/EP3618013A1/de
Publication of EP3618013A1 publication Critical patent/EP3618013A1/de
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C5/00—Registering or indicating the working of vehicles
    • G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
    • G07C5/0841—Registering performance data

Definitions

  • the invention relates to a system and a method for generating vehicle sensor data.
  • Simulation of the sensor aids in synthetically generating environment data. This reduces the effort and time required to do test drive and collect the recorded data.
  • a system for generating vehicle sensor data comprising a sensor simulation model unit, wherein the sensor simulation model unit is configured receiving synthetic vehicle ground track data comprising objects over an input interface of the sensor simulation model unit, sensing the objects by applying a sensor model of at least one sensor, wherein the sensor model comprises degradation parameters for generating degraded sensor data; and providing the degraded sensor data to an output interface of the sensor simulation model unit.
  • synthetic vehicle ground track data means artificial data in the sense that that data are created by a simulation.
  • such data relates to a composed scenario resulting from a simulation of a vehicle including its motion and position and its environment.
  • the track of the vehicle may be used.
  • real data of a recorded track e.g., a test drive may be included.
  • the output of the simulation thus is synthetic data composed of object data or graphical files.
  • the image of the graphical file may comprise, e.g., 3-D modelling data of objects in the surrounding of the vehicle, graphically created signs, etc.
  • the synthetic data is provided to the input, i.e. an input interface, of the sensor simulation model unit.
  • the data may further comprise information about the trajectory of a vehicle driving on, e.g., a road, comprising, e.g., its time dependent position and derivations thereof.
  • the data my additionally comprise information of the road scenario with traffic signs, surrounding vehicles, persons or other vehicles crossing the road, sudden events, consistence of the road including type of asphalt, holes etc.
  • the data may comprise environmental conditions as time of day, azimuth and elevation of the sun, weather and weather related road conditions, etc.
  • the sensor simulation model unit may comprise several subunits, each of which may represent a sensor of a vehicle.
  • the sensors to be simulated may be cameras, RADAR sensors, LIDAR sensors, position and navigation sensors, etc.
  • the synthetic vehicle ground track data received by the sensor simulation model unit is suitable to be "sensed" by the simulated sensors.
  • graphical data is sensed by a simulated camera, and data of objects along the road may be sensed by a simulated LIDAR or radar sensor.
  • the simulated sensors engaged in a configured scenario sense the objects as, e.g., obstacles on the road, other vehicles, etc., or images provided at the input.
  • the synthetic vehicle ground track data may be ideally simulated or already degraded.
  • a graphical image may be degraded by fog or rain.
  • the size of a detected target object by the radar sensor may be reduced due to occlusion by another object between the radar sensor and the target object.
  • the simulated sensors themselves can be configured to degrade the sensed data.
  • the sensor model comprises degradation parameters.
  • the degradation parameters comprise a set of environment degradation parameters.
  • the sensor model is extended in respect to the sensor specification parameters with regard to several parameters relating to the environment as, e.g., snow, rain, day light, etc.
  • the sensor model may be implemented a specific physical model simulating the effect on the specific sensor.
  • the set of parameters allows, e.g., to take into account environmental conditions, as, e.g., fog or rain, on one side, that degrade, e.g., the visibility of objects on an image, and on the other side may take into account effects of these environmental conditions on the hardware, which would, e.g., not be visible on an image, as, e.g., fogged lenses.
  • Further parameters may be provided to switch on/off the data transmission, to toggle bits on the data to be transmitted, to change measured values by adding a constant, a slope or random noise in order to simulate a total outage or failures of sensor.
  • the degradation parameters comprise a set of position parameters of a sensed position and/or size parameters of the detected size of a sensed object.
  • the senor provides itself position or size information of, e.g., an object to be sensed.
  • the position may be, e.g., a relative position, a distance, or a derivation thereof of an object.
  • it may also be a position provided by another sensor, as, e.g., a GNSS sensor or an inertial sensor.
  • the sensor simulation model unit is configured to associate data of at least one further sensor to the sensor.
  • positioning comprises here also the derivations as, e.g., velocity and acceleration.
  • the degradation parameters comprise a set of measurement noise parameters. Therefore, the sensor model is extended by parameters that add, e.g., Gaussian, white or colored noise or noise correlated to other parameters, as, e.g., the current light intensity or the velocity of the car.
  • the sensor model unit is configured to model a radio detection and ranging (RADAR) sensor and/or a light detection and ranging (LIDAR) sensor, a camera sensor, or further sensor types.
  • RADAR radio detection and ranging
  • LIDAR light detection and ranging
  • sensor types are positioning sensors, cameras, etc.
  • the sensors are thus capable of detecting objects to be identified, distances or derivations of distances of objects, road conditions, traffic signs, road markings, relative and absolute position information, etc.
  • a method for generating vehicle sensor data comprising the steps: receiving synthetic vehicle ground track data comprising objects over an input interface of the sensor simulation model unit, sensing the objects by applying a sensor model of at least one sensor, wherein the sensor model comprises degradation parameters for generating degraded sensor data; and providing the degraded sensor data to an output interface of the sensor simulation model unit.
  • the method reflects the functionality of the units of the system described above in the order of the dataflow, which is first receiving the synthetic vehicle ground track data at the input interface. Then, the data is processed by sensing the objects and graphical information comprised in the data and applying sensor models comprising degradation parameters in order to simulate degraded sensors.
  • the sensors may be degraded by external degradation factors, as, e.g., environmental conditions or internal degradation factors as, e.g., failure or outage of a sensor.
  • the degraded sensor data is finally output at the output interface. From there it may be transmitted for evaluation, validation and analysis to further processing modules.
  • a program element which when being executed by the processor of a sensor simulation model unit, instructs a system for generating vehicle sensor data to perform the steps of the method explained above.
  • a computer readable medium is provided on which the above-mentioned program element is stored.
  • vehicle RADAR and LIDAR sensors are widely used as sources of control signals for functionality of ADAS. These devices are operating in the millimeter wave range and infrared range in which their performance may be degraded by adverse weather conditions.
  • ADAS Advanced Driver Assistance Systems
  • millimeter-wave RADAR sensors are far less affected by adverse weather conditions than infrared based sensors.
  • automotive RADAR sensors are particularly designed for safety-oriented systems, the effects of critical issues (such as rain, fog and snow) on the sensor performance becomes of the uttermost importance. For that, system and method is provided capable to cover systematically these critical issues in tests.
  • Fig. 1 shows a system 100 for generating vehicle sensor data.
  • the system comprises a sensor simulation model unit 104, wherein the sensor simulation model unit 104 comprises the sensor simulation models 106, 108, 110 where each of the simulation models 106, 108, 110 simulates a sensor of a vehicle.
  • the system 100 is configured to receive synthetic vehicle ground track data over an input interface 102 of the sensor simulation model unit 104 and applies a sensor model 106, 108, 110 of at least one sensor to the scenario.
  • the sensor model 106, 108, 110 comprises degradation parameters for generating degraded sensor data and provides the degraded sensor data to an output interface 112 of the sensor simulation model unit 104.
  • the environment scenario around an ego vehicle could lead to different types of degradation.
  • the important scenarios of environment degradation are e.g., weather condition, occlusion and ghost target detection.
  • the degradation may result also from sensor noise.
  • the sensor is an electronic device, which may receive signals reflected from the obstacles. These signals are processed to detect and estimate object attributes. The noise in the receiver will lead to an error in the prediction of the position and orientation.
  • the main degradations to be considered are measurement errors, detection errors, position and size of the object detection errors due to occlusion and sensor failures.
  • Fig. 2 shows an embodiment of a sensor simulation model unit 202 for a sensor, which may be one of the models 106, 108, 110.
  • object parameters 204 are provided as input of the simulation model 202.
  • the objects may represent persons on or along the road, other vehicles, buildings, road markings, traffic signs, etc.
  • the input comprises further graphical data, as image files.
  • the objects and images are sensed by the sensor simulation model unit 202, and several types of degradations are applied in the sensor degradation units 208, 210, 212.
  • unit 208 may account for occlusion or other weather conditions
  • unit 210 may account for the position and size degradation and unit 212 for measurement errors.
  • the degraded data is provided to the output 206, where it may be stored in, e.g., a memory or data base for further processing or evaluation.
  • Fig. 3 shows a processing chain according to an embodiment.
  • a synthetic scenario generator tool 302 generates data 304 for real scenarios, which are acting as a ground track input data for the sensor simulation model unit 306.
  • certain real world degradation aspects can be modelled like occlusion in module 208 of Fig. 2 , position and size degradation in module 210, and measurement degradation in module 212, ghost object formation etc.
  • the output of this model is degraded data 308, which could be used by the Advanced Driver Assistance Systems (ADAS) algorithms such as Adaptive Cruise Control, Emergency Brake Assist, Lane Control functions etc. that would help in validating the ADAS functions/algorithms in a very precise and logical manner.
  • ADAS Advanced Driver Assistance Systems
  • the degraded data 308 is finally provided to the analysis, evaluation and validation units 310, 312.
  • Fig. 4 shows a method for generating vehicle sensor data according to an embodiment.
  • synthetic vehicle ground track data is received over a first interface, which may serve as an input interface.
  • Vehicle ground track data may be data of a vehicle that is moving in a scenario.
  • the trajectories, objects etc. in the scenario may be generated by a simulator or recorded during a real drive as, e.g., a test drive.
  • the data is supplied to the simulation sensor unit that comprises sensor models of at least one sensor.
  • the simulated sensors sense the objects provided at the input and apply the degradation parameters as described above.
  • the degraded sensor data is finally provided at second interface which may serve as an output interface.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
EP18191020.9A 2018-08-27 2018-08-27 System zur erzeugung von fahrzeugsensordaten Withdrawn EP3618013A1 (de)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP18191020.9A EP3618013A1 (de) 2018-08-27 2018-08-27 System zur erzeugung von fahrzeugsensordaten

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
EP18191020.9A EP3618013A1 (de) 2018-08-27 2018-08-27 System zur erzeugung von fahrzeugsensordaten

Publications (1)

Publication Number Publication Date
EP3618013A1 true EP3618013A1 (de) 2020-03-04

Family

ID=63713590

Family Applications (1)

Application Number Title Priority Date Filing Date
EP18191020.9A Withdrawn EP3618013A1 (de) 2018-08-27 2018-08-27 System zur erzeugung von fahrzeugsensordaten

Country Status (1)

Country Link
EP (1) EP3618013A1 (de)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113009900A (zh) * 2021-02-06 2021-06-22 武汉光庭信息技术股份有限公司 一种adas控制器硬件在环仿真***
US11514343B2 (en) * 2020-06-30 2022-11-29 Waymo Llc Simulating degraded sensor data

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160210775A1 (en) * 2015-01-21 2016-07-21 Ford Global Technologies, Llc Virtual sensor testbed
US9916703B2 (en) * 2015-11-04 2018-03-13 Zoox, Inc. Calibration for autonomous vehicle operation
US20180203445A1 (en) * 2017-01-13 2018-07-19 Ford Global Technologies, Llc Generating Simulated Sensor Data For Training And Validation Of Detection Models

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160210775A1 (en) * 2015-01-21 2016-07-21 Ford Global Technologies, Llc Virtual sensor testbed
US9916703B2 (en) * 2015-11-04 2018-03-13 Zoox, Inc. Calibration for autonomous vehicle operation
US20180203445A1 (en) * 2017-01-13 2018-07-19 Ford Global Technologies, Llc Generating Simulated Sensor Data For Training And Validation Of Detection Models

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MOLINA C B ET AL: "Enhancing Sensor Capabilities of Open-Source Simulation Tools to Support Autonomous Vehicles Safety Validation", 21 August 2018, INTERNATIONAL CONFERENCE ON SIMULATION, MODELING, AND PROGRAMMING FOR AUTONOMOUS ROBOTS,SIMPAR 2010; [LECTURE NOTES IN COMPUTER SCIENCE; LECT.NOTES COMPUTER], SPRINGER, BERLIN, HEIDELBERG, PAGE(S) 353 - 364, ISBN: 978-3-642-17318-9, XP047484361 *
OLAF J. GIETELINK: "Design and validation of advanced driver assistance systems", 19 November 2007 (2007-11-19), Delft, Netherlands, XP055195483, Retrieved from the Internet <URL:http://resolver.tudelft.nl/uuid:b2f0e7f6-6255-4932-8b5e-d3ef67cd81ec> [retrieved on 20150612] *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11514343B2 (en) * 2020-06-30 2022-11-29 Waymo Llc Simulating degraded sensor data
CN113009900A (zh) * 2021-02-06 2021-06-22 武汉光庭信息技术股份有限公司 一种adas控制器硬件在环仿真***

Similar Documents

Publication Publication Date Title
CN113661525B (zh) 自动驾驶车辆***和组件的仿真和验证
US10902165B2 (en) Deployable development platform for autonomous vehicle (DDPAV)
Krämmer et al. Providentia--A Large-Scale Sensor System for the Assistance of Autonomous Vehicles and Its Evaluation
US10710602B2 (en) Systems and methods for a vehicle controller safety monitor
CN111781855B (zh) 一种交通在环自动驾驶仿真***
WO2020079698A1 (en) Adas systems functionality testing
CN112146679B (zh) 用于提高自动驾驶平台的传感器i/o覆盖率的柔性测试板
CN106991041A (zh) 用于测试用于自主车辆的软件的方法和装置
EP3803646A1 (de) System und verfahren zur leistungsüberprüfung multimedischer autonomer robotersysteme
CN110356339B (zh) 一种变道盲区监测方法、***及车辆
US11198444B2 (en) Automated factory testflow of processing unit with sensor integration for driving platform
CN113508277A (zh) 车道标记定位和融合
CN114945802A (zh) 用于识别和更新自动驾驶车辆的设计适用域的***、设备和方法
US20230056233A1 (en) Sensor attack simulation system
EP3618013A1 (de) System zur erzeugung von fahrzeugsensordaten
WO2021010396A1 (ja) 走行記憶システム、潜在事故責任決定装置、走行記憶方法、潜在事故責任決定方法、映像記録システム、自動運転システム、および映像記録方法
KR20230159308A (ko) 첨단 운전자 보조 시스템(adas) 및/또는 자동화된 운전 시스템(ads)을 교정하고 검증하기 위한 방법, 시스템 및 컴퓨터 프로그램 제품
US11592810B2 (en) Systems and methods for injecting faults into an autonomy system
US11327878B2 (en) Method for rating a software component of an SiL environment
CN111103584A (zh) 用于获知车辆的环境中的物体的高度信息的设备和方法
CN113727064B (zh) 一种摄像头视场角的确定方法及装置
US20240184948A1 (en) Defining and testing evolving event sequences
WO2022219917A1 (ja) 情報処理方法及び情報処理システム
Steiner et al. Improving intelligent vehicle dependability by means of infrastructure-induced tests
US20240169111A1 (en) Parameterizing hot starts to maximize test repeatability

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN WITHDRAWN

18W Application withdrawn

Effective date: 20200623