.@Frances54371388 is the primary author of this work
Show this thread
Research on multimodal fusion using Visual-Inertial Sensor Data. We learn pose estimation using #DeepLearning in #UAV #autonomous #navigation and landing. Appearing at ACC. Details at
https://fbaldini.netlify.com/post/learning_estimation/ … @Caltechpic.twitter.com/CmqP4Yi8pM
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I may be off base here, but its twitter. So i dare ask. 1. the IMU has bias drift and will have run to run error variation. Will deep learning be better at handling those drift errors than traditional approach? and 2. does power consumption increase for Deeplearning? by how much?
-
Hi
@sunnysdesk, yes it can handle error drifts well. At first, aware of this problem, I was predicting drifting errors and correcting for it. Then, I saw that integrating IMU data directly we were able to achieve better performances. Visual features helped with the correction too
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.