Pose estimation using long range features

摘要

Aspects of the present disclosure relate to using an object detected at long range to increase the accuracy of a location and heading estimate based on near range information. For example, an autonomous vehicle may use data points collected from a sensor such as a laser to generate an environmental map of environmental features. The environmental map is then compared to pre-stored map data to determine the vehicle's geographic location and heading. A second sensor, such as a laser or camera, having a longer range than the first sensor may detect an object outside of the range and field of view of the first sensor. For example, the object may have retroreflective properties which make it identifiable in a camera image or from laser data points. The location of the object is then compared to the pre-stored map data and used to refine the vehicle's estimated location and heading.

主权项

1. A method comprising:
receiving first data collected by a first sensor while a vehicle is driving on a roadway, the first data being collected within a first sensor range of the first sensor; using the received first data from the first sensor to generate a map of the vehicle's environment within the first sensor range; estimating, by one or more processors, the vehicle's geographic location and heading by comparing the generated map to a stored map identifying the geographic locations of objects including reference objects having particular retroreflective, brightness, or intensity characteristics; receiving second data collected by a second sensor while the vehicle is driving on the roadway, the second data being collected within a second sensor range of the second sensor, wherein the second sensor range extends beyond the first sensor range; identifying a second data reference object having a particular retroreflective, brightness, or intensity characteristic from the second data collected by the second sensor, wherein the second data reference object is identified from a portion of the second data that was collected outside of the first sensor range; correlating the identified second data reference object to a reference object of the stored map based on the particular retroreflective, brightness, or intensity characteristic, the reference object of the stored map having a known geographic location; and using the known geographic location of the reference object of the stored map to refine the estimated heading of the vehicle.