beam_slam
beam_slam copied to clipboard
BEAMRobotics
Tightly coupled lidar-visual-inertial slam using the fuse framework
Reprojection constraints can skew the optimization by a lot. The goal of this is to normalize each images reproejction constraints to be in line with what would be its odometry...
Online calibration causes odometry to fail pretty frequently. Steps to debug this: 1. log the calibrations on graph updates, identify if the calibration gets skewed a lot when odom gets...
Currently, matching against the current sliding window of features doesn't result in many matches. This could be because we are using the SSC detector or because we aren't detecting enough...
The current visual path init produces a path in a bad frame and the gravity estimate is wrong as a result
Use this: https://github.com/dzunigan/imu_initialization To estimate accel bias
We want to add constraints between camera keypoints and lidar maps. We can most likely use the singleton LidarMap class in beam_slam_common, but we will likely need a way to...
Right now, the lidar odometry stores all active scans in a list, and then passes them as they come in to the scan matcher class. For the case of the...
Our current map implementatiion is not efficient for doing 3D searches. Since we have to constantly add and remove points from our map, we'd have to also rebuild a kd...
Metadata
Owner
Metadata
Tightly coupled lidar-visual-inertial slam using the fuse framework