Gaurav Pandey |
 |
Next Generation Vehicle
At Ford I was actively involved in the next generation vehicle project, a collaboration between University of Michigan, Ford Motor Company and State Farm.
For more information check out
Ford's press release and PeRL lab's NGV project page Automatic Extrinsic Calibration of Lidar and Camera
We have developed a mutual information (MI) based algorithm
for automatic extrinsic calibration of a 3D laser scanner and optical camera system.
By using MI as the registration criterion, our method is able to work in situ without
the need for any specific calibration targets, which makes it practical for in-field
calibration. More details are given in the paper "Automatic Targetless Extrinsic
Calibration of a 3D Lidar and Camera by Maximizing Mutual Information". An Information Theoretic Framework for Multi-modal Sensor Data Fusion and its Applications in Long-term Autonomous Navigation of Vehicles
We have developed an information theoretic framework for multi-modal sensor data
fusion for robust autonomous navigation of vehicles. In particular we have focused on the registra-
tion of 3D lidar and camera data, which are commonly used perception sensors in mobile
robotics. Our framework allows the fusion of the two modalities,
and uses this fused information to enhance state-of-the-art registration algorithms used in
robotics applications. The amount of information obtained
from one sensor alone is not sufficient for complex tasks like place recognition, especially
when the data in the map is significantly different from what the robot perceives in real-
time. In situations where the robot operates in changing environments we either need more
information, which can be obtained from multi-modal sensors mounted on the robot, or the
prior map needs to be updated frequently enough so that the sensor data can be registered
with data in the map. It is generally not practical to update maps of large environments
so frequently. Therefore, the ability of our algorithm to fuse multi-modal data to
recognize places across seasons, with significant appearance changes makes it very suitable
for long-term autonomous operation of robots, without the need of updating the prior map.
Three of the Ford Fusion NGV test vehicles
A simple animation to show the proposed algorithm. The top panel shows the projection
of the 3D points onto the camera image as the yaw changes. The bottom panel (right) shows the
correlation coefficient of the intensity and reflectivity values as a function of the changing yaw angle.
The bottom panel (left) shows the the joint histogram of the reflectivity and the intensity values when calculated
at these yaw angles. Note that under the correct yaw angle the correlation coefficient is maximized
and the joint histogram is least dispersed.
Sample imagery extracted from three different datasets captured in December 2009, October 2010 and February 2011; each image was captured from the same place. The datasets exhibit significant visual changes due to different weather conditions, lighting and dynamical objects.
| CONTROL & AUTOMATION / SIGNAL PROCESSING | ELECTRICAL ENGINEERING | IIT KANPUR |