SYSTEMS AND METHODS FOR LASER IMAGING ODOMETRY FOR AUTONOMOUS ROBOTS
Systems and methods for laser and imaging odometry for autonomous robots are disclosed herein. According to at least one non-limiting exemplary embodiment, a robot may utilize images captured by a sensor and encoded with a depth parameter to determine its motion and localize itself The determined motion and localization may then be utilized to verify calibration of the sensor based on a comparison between motion and localization data based on the images and motion and localization data based on data from other sensors and odometry units of the robot.
1 - 15 . (canceled)
16 . A method for determining a pose of a sensor on a robot, the method comprising:
capturing a plurality of images at different time instances using a sensing device;
determining an image disparity between the captured images by spatially transforming a subsequent image to align with an initial image;
determining a change in pose of the sensing device due to the robot moving, wherein the change in pose is determined based on the image disparity;
determining a first motion of the robot based on the determined change in pose of the sensing device;
determining a second motion of the robot based on data from at least one interoceptive sensor, wherein the second motion is determined using a probabilistic distribution of the pose of the robot;
determining the pose of the sensing device based on a motion disparity between the first motion determined from the change in pose of the sensing device and the second motion of the robot; and
localizing an object on a computer readable map based at least in part on a most probable pose of the sensing device.
17 . The method of claim 16 , further comprising updating the probabilistic distribution of the pose of the robot based on the determined pose of the sensing device.
18 . The method of claim 16 , wherein determining the second motion of the robot comprises:
representing the pose of the robot with a plurality of particles, each particle having a position and an orientation; and
updating the positions and orientations of the particles based on the data from the at least one interoceptive sensor.
19 . The method of claim 18 , wherein updating the positions and orientations of the particles is performed sequentially.
20 . The method of claim 16 , wherein capturing the plurality of images comprises capturing a first image at an initial time instance and capturing a second image at a subsequent time instance using the sensing device, wherein the first image and the second image comprise pixels of at least one target object, and
21 . The method of claim 20 , further comprising determining the change in pose of the sensing device based on spatial transformations applied to the subsequent image.
22 . The method of claim 16 , wherein the sensing device comprises at least one of a stereo camera or a depth sensor.
23 . The method of claim 16 , wherein the at least one interoceptive sensor comprises at least one of an inertial measurement unit (IMU), such as encoders, gyroscopes, or accelerometers.
24 . The method of claim 16 , further comprising:
determining an expected measurement of at least one target object at an expected position of the robot based on odometry data; and
determining an actual position of the robot based on a difference between the expected measurement and an actual measurement of the at least one target object obtained from the captured images.
25 . The method of claim 24 , wherein the actual position of the robot is determined with improved accuracy by accounting for wheel slippage experienced by the robot during navigation.
26 . A system for determining a pose of a sensor on a robot, the system comprising:
a sensing device configured to capture a plurality of images at different time instances; and
a processor configured to execute computer readable instructions to:
determine an image disparity between the captured images by spatially transforming a subsequent image to align with an initial image;
determine a change in pose of the sensing device due to the robot moving, wherein the change in pose is determined based on the image disparity;
determine a motion of the robot based on the determined change in pose of the sensing device;
determine a second motion of the robot based on data from at least one interoceptive sensor, wherein the second motion is determined using a probabilistic distribution of the pose of the robot;
determine the pose of the sensing device based on a motion disparity between the motion determined from the change in pose of the sensing device and the second motion of the robot; and
localize an object on a computer readable map based at least in part on a most probable pose of the sensing device.
27 . The system of claim 26 , wherein the processor is further configured to execute the computer readable instructions to update the probabilistic distribution of the pose of the robot based on the determined pose of the sensing device.
28 . The system of claim 26 , wherein determining the second motion of the robot comprises:
representing the pose of the robot with a plurality of particles, each particle having a position and an orientation; and
updating the positions and orientations of the particles based on the data from the at least one interoceptive sensor.
29 . The system of claim 28 , wherein updating the positions and orientations of the particles is performed sequentially.
30 . The system of claim 26 , wherein capturing the plurality of images comprises capturing a first image at an initial time instance and capturing a second image at a subsequent time instance using the sensing device, wherein the first image and the second image comprise pixels of at least one target object.
31 . The system of claim 30 , wherein the processor is further configured to execute the computer readable instructions to,
determine the change in pose of the sensing device based on spatial transformations applied to the subsequent image.
32 . The system of claim 26 , wherein the sensing device comprises at least one of a stereo camera or a depth sensor, and wherein the at least one interoceptive sensor comprises at least one of an inertial measurement unit (IMU), such as encoders, gyroscopes, or accelerometers.
33 . The system of claim 26 , wherein the processor is further configured to execute the computer readable instructions to:
determine an expected measurement of at least one target object at an expected position of the robot based on odometry data; and
determine an actual position of the robot based on a difference between the expected measurement and an actual measurement of the at least one target object obtained from the captured images.
34 . The system of claim 33 , wherein the actual position of the robot is determined with improved accuracy by accounting for wheel slippage experienced by the robot during navigation.