CNN gate detection (CORRECTED)#122
Open
tommasocalzolari wants to merge 1 commit into
Open
Conversation
Closed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The previous Pull Request for CNN gate detection had several issues because of problems with git, this new pull request should be clean and working correctly. Please disregard the previous Pull Request with the same name and sorry for the confusion. (Only 1 commit is visible because we had to recreate the solution from a new clean branch)
GROUP 9
This pull request integrates a lightweight CNN-based gate detector for TU Delft MAVLab gates into Paparazzi and connects it to an initial navigation module for Bebop-based experiments.
The detector is implemented as a Paparazzi computer-vision module and runs fully onboard in C, without relying on external machine-learning libraries during inference. The network was trained offline on images of gates, then validated and tested before deployment. The trained weights were embedded directly in the source tree so that onboard computation is limited to image preprocessing and forward inference.
In addition to the detector, this pull request wires the module into a dedicated custom airframe and a course configuration entry, and it connects the detector output to custom_avoider, which acts as an initial gate-oriented navigation concept. That navigation module is intentionally a first integration step and still requires further refinement, tuning, and validation before being considered a robust autonomous solution.