Simultaneous localization and mapping for a mobile robot
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G05D-001/00
G05D-001/02
출원번호
US-0674427
(2015-03-31)
등록번호
US-9400501
(2016-07-26)
발명자
/ 주소
Schnittman, Mark
출원인 / 주소
iRobot Corporation
대리인 / 주소
Honigman Miller Schwartz and Cohn LLP
인용정보
피인용 횟수 :
0인용 특허 :
63
초록▼
A method of simultaneous localization and mapping includes initializing a robot pose and a particle model of a particle filter. The particle model includes particles, each having an associated map, robot pose, and weight. The method includes receiving sparse sensor data from a sensor system of the r
A method of simultaneous localization and mapping includes initializing a robot pose and a particle model of a particle filter. The particle model includes particles, each having an associated map, robot pose, and weight. The method includes receiving sparse sensor data from a sensor system of the robot, synchronizing the received sensor data with a change in robot pose, accumulating the synchronized sensor data over time, and determining a robot localization quality. When the accumulated sensor data exceeds a threshold accumulation and the robot localization quality is greater than a threshold localization quality, the method includes updating particles with accumulated synchronized sensor data. The method includes determining a weight for each updated particle of the particle model and setting a robot pose belief to the robot pose of the particle having the highest weight when a mean weight of the particles is greater than a threshold particle weight.
대표청구항▼
1. A method comprising: receiving, at a computing device, sensor data from an autonomous mobile robot operating in a work environment;updating, using the computing device, an occupancy map of the work environment with location occupancy probabilities based on the received sensor data;determining, us
1. A method comprising: receiving, at a computing device, sensor data from an autonomous mobile robot operating in a work environment;updating, using the computing device, an occupancy map of the work environment with location occupancy probabilities based on the received sensor data;determining, using the computing device, a localization quality of the autonomous mobile robot; andwhen the localization quality does not satisfy a threshold localization quality, executing, using the computing device, a wall-following behavior that causes the autonomous mobile robot to maneuver toward a wall in the work environment and drive adjacent to the wall until the localization quality satisfies the threshold localization quality. 2. The method of claim 1, wherein the wall-following behavior uses proximity sensor data from a proximity sensor of the autonomous mobile robot to cause the autonomous mobile robot to maintain a threshold distance from the wall while driving adjacent the wall. 3. The method of claim 1, further comprising, when the localization quality does not satisfy the threshold localization quality, refraining from updating the occupancy map. 4. The method of claim 1, further comprising, when the localization quality does not satisfy the threshold localization quality: identifying a location of the wall using the occupancy map; andissuing a drive command from the computing device to a drive system of the autonomous mobile robot to maneuver the autonomous mobile robot toward the identified location of the wall. 5. The method of claim 4, further comprising: after updating the occupancy map, storing the updated occupancy map in memory storage hardware in communication with the computing device; andwhen the localization quality does not satisfy the threshold localization quality: retrieving a previously stored occupancy map from the memory storage hardware; andidentifying the location of the wall using the previously stored occupancy map. 6. The method of claim 1, further comprising: synchronizing the received sensor data with a change in robot pose;accumulating the synchronized sensor data over time in memory storage hardware in communication with the computing device; andwhen the received sensor data exceeds a threshold accumulation and the localization quality satisfies the threshold localization quality, updating particles of a particle model of a particle filter with the accumulated synchronized sensor data, each particle having an associated particle occupancy map, robot pose, and weight. 7. The method of claim 6, wherein the sensor data comprises range data from at least one range finding sensor of the autonomous mobile robot and the synchronized sensor data comprises range measurement and bearing pairs. 8. The method of claim 7, wherein the threshold accumulation of the sensor data comprises at least 20 range measurements from the at least one range finding sensor. 9. The method of claim 6, wherein determining the robot localization quality comprises determining an average weight of the particles of the particle model. 10. The method of claim 6, further comprising: determining the weight for each updated particle of the particle model based on whether the received sensor data corroborates the corresponding particle occupancy map; andsetting a robot pose belief to the robot pose of the particle having a highest weight when a mean weight of the particles is greater than a threshold particle weight. 11. The method of claim 6, further comprising, when the localization quality does not satisfy the threshold localization quality, adding new particles to the particle model based on the sensor data received while executing the wall-following behavior. 12. The method of claim 6, further comprising: sampling the particle model for a particle population; anddetermining, using the computing device, an instantaneous localization of the autonomous mobile robot based on at least one particle of the particle population. 13. The method of claim 12, further comprising resampling the particle model for the particle population when the localization quality satisfies the threshold localization quality. 14. The method of claim 12, further comprising, when the localization quality does not satisfy the threshold localization quality, replacing a portion of the particle population with particles selected from the particle model based on the accumulated synchronized sensor data. 15. The method of claim 1, wherein determining the robot localization quality comprises: computing an instantaneous localization based on the sensor data;receiving the instantaneous localization through a fast low-pass filter;receiving the instantaneous localization through a slow low-pass filter; andreceiving the instantaneous localization through a leaky integrator;wherein the localization quality does not satisfy the threshold localization quality when an integrator value of the leaky integrator fails to satisfy an integrator threshold. 16. A method comprising: receiving, at a computing device, sensor data from an autonomous mobile robot operating in a work environment;updating, using the computing device, an occupancy map of the work environment with location occupancy probabilities based on the received sensor data;determining, using the computing device, a localization quality of the autonomous mobile robot; andwhen the localization quality does not satisfy a threshold localization quality: identifying, using the computing device, a mapped object on the occupancy map; andissuing a drive command from the computing device to a drive system of the autonomous mobile robot to maneuver the autonomous mobile robot toward the mapped object in the work environment to re-localize the autonomous mobile robot based on the sensor data. 17. The method of claim 16, further comprising: comparing, using the computing device, the sensor data relative to the mapped object on the occupancy map; anddetermining, using the computing device, a pose belief of the autonomous mobile robot based on the comparison. 18. The method of claim 16, further comprising, when the localization quality does not satisfy the threshold localization quality, refraining from updating the occupancy map. 19. The method of claim 16, further comprising: after updating the occupancy map, storing the updated occupancy map in memory storage hardware in communication with the computing device; andwhen the localization quality does not satisfy the threshold localization quality: retrieving a previously stored occupancy map from the memory storage hardware; andissuing the drive command based on the previously stored occupancy map. 20. The method of claim 16, further comprising: synchronizing the received sensor data with a change in robot pose;accumulating the synchronized sensor data over time in memory storage hardware in communication with the computing device; andwhen the received sensor data exceeds a threshold accumulation and the localization quality satisfies the threshold localization quality, updating particles of a particle model of a particle filter with the accumulated synchronized sensor data, each particle having an associated particle occupancy map, robot pose, and weight. 21. The method of claim 20, wherein the sensor data comprises range data from at least one range finding sensor of the autonomous mobile robot and the synchronized sensor data comprises range measurement and bearing pairs. 22. The method of claim 21, wherein the threshold accumulation of the sensor data comprises at least 20 range measurements from the at least one range finding sensor. 23. The method of claim 20, wherein determining the robot localization quality comprises determining an average weight of the particles of the particle model. 24. The method of claim 20, further comprising: determining the weight for each updated particle of the particle model based on whether the received sensor data corroborates the corresponding particle occupancy map; andsetting a robot pose belief to the robot pose of the particle having a highest weight when a mean weight of the particles is greater than a threshold particle weight. 25. The method of claim 20, further comprising, when the localization quality does not satisfy the threshold localization quality, adding new particles to the particle model based on the sensor data received while the autonomous mobile robot maneuvers toward the mapped object in the work environment. 26. The method of claim 20, further comprising: sampling the particle model for a particle population; anddetermining, using the computing device, an instantaneous localization of the autonomous mobile robot based on at least one particle of the particle population. 27. The method of claim 26, further comprising resampling the particle model for the particle population when the localization quality satisfies the threshold localization quality. 28. The method of claim 26, further comprising, when the localization quality does not satisfy the threshold localization quality, replacing a portion of the particle population with particles selected from the particle model based on the accumulated synchronized sensor data. 29. The method of claim 16, wherein determining the robot localization quality comprises: computing an instantaneous localization based on the sensor data;receiving the instantaneous localization through a fast low-pass filter;receiving the instantaneous localization through a slow low-pass filter; andreceiving the instantaneous localization through a leaky integrator;wherein the localization quality does not satisfy the threshold localization quality when an integrator value of the leaky integrator fails to satisfy an integrator threshold.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (63)
Lanckton Arnold H. (Roma NY) More Randall K. (Manlius NY), Advanced terrain mapping system.
Steffens Johannes Bernhard ; Elagin Egor Valerievich ; Nocera Luciano Pasquale Agostino ; Maurer Thomas ; Neven Hartmut, Face recognition from video images.
Eric Richard Bartsch ; Charles William Fisher ; Paul Amaat France ; James Frederick Kirkpatrick ; Gary Gordon Heaton ; Thomas Charles Hortel ; Arseni Velerevich Radomyselski ; James Randy Stig, Home cleaning robot.
Tuck Alan,GBX ; Brayson Gary,GBX ; Ignagni Mario B. ; Touchberry Alan B. ; Anderson Donald William,GBX ; Glen Stephen James,GBX ; Gilman James Michael Alexander,GBX, Method and apparatus for determining location of characteristics of a pipeline.
Shimano,Mihoko; Nagao,Kenji; Akimoto,Toshiaki; Naruoka,Tomonobu, Method and apparatus for object recognition using a plurality of cameras and databases.
Caminiti, Lorenzo; Goncalves, Luis; Di Bernardo, Enrico; Moursund, Carter, Method and system for automatically determining lines of sight between nodes.
McGee H. Dean (Rochester Hills MI) Krause Kenneth W. (Rochester MI) Coldren Bruce E. (Troy MI), Method and system for automatically determining the position and orientation of an object in 3-D space.
Asaka Shunichi (Sagamihara JPX) Echigo Tomio (Yokohama JPX) Hazeki Shinichiro (Kawasaki JPX) Ishikawa Shigeki (Tokyo JPX), Method and system for maneuvering a mobile robot.
Hanna Keith J. (Princeton NJ) Kumar Rakesh (Dayton NJ), Method for estimating the location of an image target region from tracked multiple image landmark regions.
Bauer Rudolf (Neubiberg DEX), Method for producing a cellularly structured environment map of a self-propelled, mobile unit that orients itself in the.
McTamaney Louis S. (Cupertino CA) Wong Yue M. (Saratoga CA) Chandra Rangasami S. (Pleasanton CA) Walker Robert A. (Sunnyvale CA) Lastra Jorge E. (San Jose CA) Wagner Paul A. (Cambridge MA) Sharma Uma, Multi-purpose autonomous vehicle with path plotting.
Everett ; Jr. Hobart R. (San Diego CA) Gilbreath Gary A. (San Diego CA) Laird Robin T. (San Diego CA), Navigational control system for an autonomous vehicle.
Goncalves,Luis Filipe Domingues; Di Bernardo,Enrico; Pirjanian,Paolo; Karlsson,L. Niklas, Systems and methods for computing a relative pose for global localization in a visual simultaneous localization and mapping system.
Goncalves,Luis Filipe Domingues; Karlsson,L. Niklas; Pirjanian,Paolo; Di Bernardo,Enrico, Systems and methods for controlling a density of visual landmarks in a visual simultaneous localization and mapping system.
Karlsson,L. Niklas; Goncalves,Luis Filipe Domingues; Di Bernardo,Enrico; Pirjanian,Paolo, Systems and methods for correction of drift via global localization with a visual landmark.
Goncalves,Luis Filipe Domingues; Karlsson,L. Niklas; Pirjanian,Paolo; Di Bernardo,Enrico, Systems and methods for filtering potentially unreliable visual data for visual simultaneous localization and mapping.
Karlsson,L. Niklas; Pirjanian,Paolo; Goncalves,Luis Filipe Domingues; Di Bernardo,Enrico, Systems and methods for incrementally updating a pose of a mobile device calculated by visual simultaneous localization and mapping techniques.
Domingues Goncalves, Luis Filipe; Di Bernardo, Enrico; Pirjanian, Paolo; Karlsson, L. Niklas, Systems and methods for landmark generation for visual simultaneous localization and mapping.
Karlsson,L. Niklas; Pirjanian,Paolo; Goncalves,Luis Filipe Domingues; Di Bernardo,Enrico, Systems and methods for using multiple hypotheses in a visual simultaneous localization and mapping system.
Evans ; Jr. John M. (Brookfield CT) Weiman Carl F. R. (Westport CT) King Steven J. (Woodbury CT), Visual navigation and obstacle avoidance structured light system.
Evans ; Jr. John M. (Brookfield CT) Weiman Carl F. R. (Westport CT) King Steven J. (Woodbury CT), Visual navigation and obstacle avoidance structured light system.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.