A method and apparatus are provided for optimizing one or more object detection parameters used by an autonomous vehicle to detect objects in images. The autonomous vehicle may capture the images using one or more sensors. The autonomous vehicle may then determine object labels and their correspondi
A method and apparatus are provided for optimizing one or more object detection parameters used by an autonomous vehicle to detect objects in images. The autonomous vehicle may capture the images using one or more sensors. The autonomous vehicle may then determine object labels and their corresponding object label parameters for the detected objects. The captured images and the object label parameters may be communicated to an object identification server. The object identification server may request that one or more reviewers identify objects in the captured images. The object identification server may then compare the identification of objects by reviewers with the identification of objects by the autonomous vehicle. Depending on the results of the comparison, the object identification server may recommend or perform the optimization of one or more of the object detection parameters.
대표청구항▼
1. An apparatus for optimizing object detection performed by an autonomous vehicle, the apparatus comprising: a memory operative to store: a plurality of images captured by an autonomous vehicle using a plurality of object detection parameters;a first plurality of object label parameters determined
1. An apparatus for optimizing object detection performed by an autonomous vehicle, the apparatus comprising: a memory operative to store: a plurality of images captured by an autonomous vehicle using a plurality of object detection parameters;a first plurality of object label parameters determined by the autonomous vehicle; anda second plurality of object label parameters applied by a reviewer having reviewed the plurality of images captured by the autonomous vehicle, the first and second pluralities of object label parameters defining a shape bounding a detected object; anda processor in communication with the memory, the processor operative to: determine whether to optimize the plurality of object detection parameters based on a comparison of the first plurality of object label parameters with the second plurality of object label parameters; andperform an operation on the plurality of object detection parameters based on the comparison of the first plurality of object label parameters with the second plurality of object label parameters, wherein the operation comprises: identifying a plurality of object detection values, wherein each object detection value corresponds to at least one object detection parameter in the plurality of object detection parameters;for each combination of the plurality of object detection values, performing an object detection routine on the plurality of images captured by the autonomous vehicle using the plurality of object detection values; andselecting the combination of the plurality of object detection values that resulted in an optimal object detection routine. 2. The apparatus of claim 1, wherein the operation further comprises displaying the combination of the plurality of object detection values that resulted in an optimal object detection routine. 3. The apparatus of claim 1, wherein a number of times the object detection routine is performed is equal to a number of object detection parameters of the plurality of object detection parameters raised to a power of a number the plurality of the object detection values. 4. The apparatus of claim 3, wherein the number of times the object detection routine is performed equals a same number of different sets of the first plurality of object label parameters stored in memory. 5. The apparatus of claim 1, wherein the plurality of object detection values includes different sensor types used for images captured by the autonomous vehicle. 6. The apparatus of claim 1, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining whether the first plurality of object label parameters overlaps with any portion of the second plurality of object label parameters. 7. The apparatus of claim 1, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining whether the autonomous vehicle has met a predetermined missed object threshold corresponding to a maximum allowable quantity of the first plurality of object label parameters not applied by the autonomous vehicle compared to a threshold value. 8. An method for optimizing object detection performed by an autonomous vehicle, the method comprising: storing, in a memory, a plurality of images captured by an autonomous vehicle using a plurality of object detection parameters,receiving, by one or more computing devices having one or more processors, a first plurality of object label parameters determined by the autonomous vehicle and a second plurality of object label parameters applied by a reviewer having reviewed the plurality of images captured by the autonomous vehicle, the first and second pluralities of object label parameters defining a shape bounding a detected object;determining, by the one or more computing devices, whether to optimize the plurality of object detection parameters based on a comparison of the first plurality of object label parameters with the second plurality of object label parameters;performing, by the one or more computing devices, an operation on the plurality of object detection parameters based on the comparison of the first plurality of object label parameters with the second plurality of object label parameters;identifying, by the one or more computing devices, a plurality of object detection values, wherein each object detection value corresponds to at least one object detection parameter in the plurality of object detection parameters;performing, by the one or more computing devices, an object detection routine on the plurality of images captured by the autonomous vehicle using the plurality of object detection values for each combination of the plurality of object detection values; andselecting, by the one or more computing devices, the combination of the plurality of object detection values that resulted in an optimal object detection routine. 9. The method of claim 8, further comprising displaying, by the one or more computing devices, the combination of the plurality of object detection values that resulted in an optimal object detection routine. 10. The method of claim 8, wherein a number of times the object detection routine is performed is equal to a number of object detection parameters of the plurality of object detection parameters raised to a power of a number the plurality of the object detection values. 11. The method of claim 10, wherein the number of times the object detection routine is performed equals a same number of different sets of the first plurality of object label parameters stored in memory. 12. The method of claim 8, wherein the plurality of object detection values includes different sensor types used for images captured by the autonomous vehicle. 13. The method of claim 8, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining, by the one or more computing devices, whether the first plurality of object label parameters overlaps with any portion of the second plurality of object label parameters. 14. The method of claim 8, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining, by the one or more computing devices, whether the autonomous vehicle has met a predetermined missed object threshold corresponding to a maximum allowable quantity of the first plurality of object label parameters not applied by the autonomous vehicle compared to a threshold value. 15. A non-transitory, tangible recording medium on which instructions are stored, the instructions, when executed by one or more processors of one or more computing devices, causes the one or more processor to perform a method for optimizing object detection performed by an autonomous vehicle, the method comprising: storing a plurality of images captured by an autonomous vehicle using a plurality of object detection parameters,receiving a first plurality of object label parameters determined by the autonomous vehicle and a second plurality of object label parameters applied by a reviewer having reviewed the plurality of images captured by the autonomous vehicle;determining whether to optimize the plurality of object detection parameters based on a comparison of the first plurality of object label parameters with the second plurality of object label parameters;performing an operation on the plurality of object detection parameters based on the comparison of the first plurality of object label parameters with the second plurality of object label parameters, the first and second pluralities of object label parameters defining a shape bounding a detected object;identifying a plurality of object detection values, wherein each object detection value corresponds to at least one object detection parameter in the plurality of object detection parameters;performing an object detection routine on the plurality of images captured by the autonomous vehicle using the plurality of object detection values for each combination of the plurality of object detection values; andselecting the combination of the plurality of object detection values that resulted in an optimal object detection routine. 16. The medium of claim 15, further comprising displaying the combination of the plurality of object detection values that resulted in an optimal object detection routine. 17. The medium of claim 15, wherein a number of times the object detection routine is performed is a number of object detection parameters of the plurality of object detection parameters raised to a power of a number the plurality of the object detection values. 18. The medium of claim 17, wherein the number of times the object detection routine is performed equals a same number of different sets of the first plurality of object label parameters stored in a memory. 19. The medium of claim 15, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining whether the first plurality of object label parameters overlaps with any portion of the second plurality of object label parameters. 20. The medium of claim 15, wherein the comparison of the first plurality of object label parameters with the second plurality of object label parameters includes determining whether the autonomous vehicle has met a predetermined missed object threshold corresponding to a maximum allowable quantity of the first plurality of object label parameters not applied by the autonomous vehicle compared to a threshold value.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (115)
Oda, Tamami; Watanabe, Tomo; Sato, Tsuyoshi, AUTOMATIC VEHICLE GUIDANCE SYSTEM, CONTROL APPARATUS IN AUTOMATIC VEHICLE GUIDANCE SYSTEM, AUTOMATIC VEHICLE GUIDANCE METHOD, AND COMPUTER-READABLE DATA RECORDED MEDIUM IN WHICH AUTOMATIC VEHICLE GUI.
Huang, Jihua; Lin, William C.; Chin, Yuen-Kwok, Adaptive vehicle control system with driving style recognition based on maneuvers at highway on/off ramps.
Isogai Akira,JPX ; Takagi Kiyokazu,JPX, Automatic deceleration control system, vehicle-to-obstacle distance control system, and system program storage medium for vehicle.
Latarnik Michael (Friedrichsdorf DEX) Kolbe Alexander (Gross Zimmern DEX) Honus Klaus (Frankfurt am Main DEX), Circuit configuration and method for controlling a traction slip control system with brake and/or engine management.
Trepagnier, Paul Gerard; Nagel, Jorge Emilio; Dooner, Matthew Taylor; Dewenter, Michael Thomas; Traft, Neil Michael; Drakunov, Sergey; Kinney, Powell; Lee, Aaron, Control and systems for autonomously driven vehicles.
Trepagnier, Paul Gerard; Nagel, Jorge Emilio; Dooner, Matthew Taylor; Dewenter, Michael Thomas; Traft, Neil Michael; Drakunov, Sergey; Kinney, Powell; Lee, Aaron, Control and systems for autonomously driven vehicles.
Trepagnier, Paul Gerard; Nagel, Jorge Emilio; Dooner, Matthew Taylor; Dewenter, Michael Thomas; Traft, Neil Michael; Drakunov, Sergey; Kinney, Powell; Lee, Aaron, Control and systems for autonomously driven vehicles.
Wimmer, Martin; Meurle, Juergen; Sacher, Heike; Siedersberger, Karl-Heinz; Scheiffert, Fabian; Ehrhart, Franziska, Device for the automated driving of a motor vehicle, motor vehicle having such a device and method for operating a motor vehicle.
Brett B. Stewart ; James W. Thompson, Geographic-based communication service system with more precise determination of a user's known geographic location.
Holloway,Lane Thomas; Kobrosly,Walid M.; Malik,Nadeem; Quiller,Marques Benjamin, Limiting and controlling motor vehicle operation functions enabled for each of a group of drivers of the vehicle.
Niem, Wolfgang; Ibele, Philipp, Method and device for recognizing road signs in the vicinity of a vehicle and for synchronization thereof to road sign information from a digital map.
Christopher Kenneth Hoover Wilson ; Seth Olds Rogers ; Patrick Wyatt Langley, Method and system for autonomously developing or augmenting geographical databases by mining uncoordinated probe data.
Bhogal, Kulvir S.; Peterson, Robert R.; Seacat, Lisa A.; Talbot, Mark W., Method and system to alert user of local law via the Global Positioning System (GPS).
Toepfer Bernhard (Stuttgart DEX) Reiner Michael (Fellbach DEX) Klein Bodo (Borsinghausen DEX), Method of determining a wear-dependent braking force distribution.
Trepagnier, Paul Gerard; Nagel, Jorge Emilio; Kinney, Powell McVay; Dooner, Matthew Taylor; Wilson, Bruce Mackie; Schneider, Jr., Carl Reimers; Goeller, Keith Brian, Navigation and control system for autonomous vehicles.
O Connor, Michael L.; Bell, Thomas; Eglington, Michael L.; Leckie, Lars; Gutt, Gregory M.; Zimmerman, Kurt R., Rapid adjustment of trajectories for land vehicles.
Rao Prithvi N. (Pittsburgh PA) Shin Dong Hun (Pittsburgh PA) Whittaker William L. (Pittsburgh PA) Kleimenhagen Karl W. (Peoria IL) Singh Sanjiv J. (Pittsburgh PA) Kemner Carl A. (Peoria Heights IL) B, System and method for enabling an autonomous vehicle to track a desired path.
Barrett,David S.; Allard,James; Filippov,Misha; Pack,Robert Todd; Svendsen,Selma, System and method for processing safety signals in an autonomous vehicle.
Norris, William Robert; Allard, James; Filippov, Mikhail O.; Haun, Robert Dale; Turner, Christopher David Glenn; Gilbertson, Seth; Norby, Andrew Julian, Systems and methods for switching between autonomous and manual operation of a vehicle.
Nagda, Paresh; Li, Wenbin; Howlett, Julia; Fan, Rodric C.; Yang, Xinnong; Fay, James D., Using location data to determine traffic and route information.
Harumoto,Satoshi; Yamato,Toshitaka; Takeuchi,Hiroshi; Maeno,Yoshihiko; Miyamoto,Naotoshi; Sakiyama,Kazuhiro, Vehicle control apparatus, vehicle control method, and computer program.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.