Velocity estimation and object tracking for autonomous vehicle applications
View Patent ↗Aspects and implementations of the present disclosure address shortcomings of the existing technology by enabling velocity estimation for efficient object identification and tracking in autonomous vehicle (AV) applications, including: obtaining, by a sensing system of the AV, a plurality of return points, each return point having a velocity value and coordinates of a reflecting region that reflects a signal emitted by the sensing system, identifying an association of the velocity values and the coordinates of return points with a motion of a physical object, the motion being a combination of a translational motion and a rotational motion of a rigid body, and causing a driving path of the AV to be determined in view of the motion of the physical object.
1 . A method comprising:
obtaining, by a sensing system of an autonomous vehicle, a plurality of return points, each return point comprising a radial velocity value and one or more coordinates of a reflecting region that reflects a signal emitted by the sensing system;
associating a first subset of the plurality of return points with one or more characteristics of a motion of an object, the one or more characteristics comprising a combination of a rotational velocity of the object and a first component of a translational velocity of the object;
applying a disambiguation scheme to the combination to disambiguate the rotational velocity from the first component of the translational velocity; and
causing a driving path of the autonomous vehicle to be modified in view of the one or more characteristics of the motion of the object.
2 . The method of claim 1 , wherein applying the disambiguation scheme comprises:
using a road curvature to associate the translational velocity of the object with the rotational velocity of the object.
3 . The method of claim 1 , wherein applying the disambiguation scheme comprises:
obtaining an additional combination of the rotational velocity of the object and the first component of the translational velocity of the object, wherein the additional combination is obtained by associating a second subset of the plurality of return points with the one or more characteristics of the motion of the object, the first subset of the plurality of return points and the second subset of the plurality of return points obtained for different times.
4 . The method of claim 3 , wherein the first subset of the plurality of return points and the second subset of the plurality of return points correspond to different cycles of a transmitter of the sensing system.
5 . The method of claim 1 , wherein applying the disambiguation scheme comprises:
estimating the first component of the translational velocity of the object based on a first image of the object at a first time and a second image of the object at a second time.
6 . The method of claim 1 , wherein associating the first subset of the plurality of return points with the one or more characteristics of the object comprises:
using the one or more characteristics to fit the first subset of the plurality of return points to a rigid-body motion of the object.
7 . The method of claim 6 , further comprising:
evaluating a degree of fitting of the one or more characteristics to the first subset of the plurality of return points using a loss function that aggregates a plurality of loss values, an individual loss value of the plurality of loss values characterizing a mismatch between a corresponding return point of the first subset of the plurality of return points and the rigid-body motion of the object.
8 . The method of claim 1 , wherein the plurality of return points are obtained using a lidar sensor of the sensing system of the autonomous vehicle.
9 . The method of claim 1 , further comprising:
tracking the object across an interval of times using at least the first component of the translational velocity of the object and a second component of the translational velocity of the object, wherein the first component of the translational velocity is associated with a lateral part of the motion of the object and the second component of the translational velocity is associated with a radial part of the motion of the object.
10 . A system comprising:
a sensing system of an autonomous vehicle, the sensing system to:
obtain a plurality of return points, each return point comprising a radial velocity value and one or more coordinates of a reflecting region that reflects a signal emitted by the sensing system; and
a perception system of the autonomous vehicle, the perception system to:
associate a first subset of the plurality of return points with one or more characteristics of a motion of an object, the one or more characteristics comprising a combination of a rotational velocity of the object and a first component of a translational velocity of the object;
apply a disambiguation scheme to the combination to disambiguate the rotational velocity from the first component of the translational velocity; and
cause a driving path of the autonomous vehicle to be modified in view of the one or more characteristics of the motion of the object.
11 . The system of claim 10 , wherein to apply the disambiguation scheme, the perception system of the autonomous vehicle is to:
use a road curvature to associate the translational velocity of the object with the rotational velocity of the object.
12 . The system of claim 10 , wherein to apply the disambiguation scheme, the perception system of the autonomous vehicle is to:
obtain an additional combination of the rotational velocity of the object and the first component of the translational velocity of the object, wherein the additional combination is obtained by associating a second subset of the plurality of return points with the one or more characteristics of the motion of the object, the first subset of the plurality of return points and the second subset of the plurality of return points obtained for different times.
13 . The system of claim 12 , wherein the first subset of the plurality of return points and the second subset of the plurality of return points correspond to different cycles of a transmitter of the sensing system.
14 . The system of claim 10 , wherein to apply the disambiguation scheme, the perception system of the autonomous vehicle is to:
estimate the first component of the translational velocity of the object based on a first image of the object at a first time and a second image of the object at a second time.
15 . The system of claim 10 , wherein to associate the first subset of the plurality of return points with the one or more characteristics of the object, the perception system of the autonomous vehicle is to:
use the one or more characteristics to fit the first subset of the plurality of return points to a rigid-body motion of the object.
16 . The system of claim 15 , wherein the perception system of the autonomous vehicle is further to:
evaluate a degree of fitting of the one or more characteristics to the first subset of the plurality of return points using a loss function that aggregates a plurality of loss values, an individual loss value of the plurality of loss values characterizing a mismatch between a corresponding return point of the first subset of the plurality of return points and the rigid-body motion of the object.
17 . The system of claim 10 , wherein the plurality of return points are obtained using a lidar sensor of the sensing system of the autonomous vehicle.
18 . The system of claim 10 , wherein the perception system of the autonomous vehicle is further to:
track the object across an interval of times using at least the first component of the translational velocity of the object and a second component of the translational velocity of the object, wherein the first component of the translational velocity is associated with a lateral part of the motion of the object and the second component of the translational velocity is associated with a radial part of the motion of the object.
19 . An autonomous vehicle comprising:
a sensing system configured to:
obtain a plurality of return points, each return point comprising a radial velocity value and one or more coordinates of a reflecting region that reflects a signal emitted by the sensing system; and
a perception system configured to:
associate a first subset of the plurality of return points with one or more characteristics of a motion of an object, the one or more characteristics comprising a combination of a rotational velocity of the object and a first component of a translational velocity of the object;
apply a disambiguation scheme to the combination to disambiguate the rotational velocity from the first component of the translational velocity; and
cause a driving path of the autonomous vehicle to be modified in view of the one or more characteristics of the motion of the object.
20 . The autonomous vehicle of claim 19 , wherein to apply the disambiguation scheme, the perception system is configured to perform at least one of:
use a road curvature to associate the translational velocity of the object with the rotational velocity of the object;
obtain an additional combination of the rotational velocity of the object and the first component of the translational velocity of the object, wherein the additional combination is obtained by associating a second subset of the plurality of return points with the one or more characteristics of the motion of the object, the first subset of the plurality of return points and the second subset of the plurality of return points obtained for different times; or
estimate the first component of the translational velocity of the object based on a first image of the object at a first time and a second image of the object at a second time.