From 76dbd7cf45e275caea141d5f09f4b7786f2d3171 Mon Sep 17 00:00:00 2001 From: hollydinkel Date: Mon, 25 Mar 2024 09:23:42 -0700 Subject: [PATCH] [DOCS] minor changes to data_processing.md, still WIP --- docs/data_processing.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/data_processing.md b/docs/data_processing.md index fe5f0aa..61f3ead 100644 --- a/docs/data_processing.md +++ b/docs/data_processing.md @@ -2,13 +2,13 @@ The files documents the steps for creating a dataset to use with Fast Change Detection. -First, install the [Astrobee flight software](https://github.com/nasa/astrobee) in the '$HOME' directory. Additional instructions for using the Astrobee flight software with Docker are [here](https://docs.google.com/document/d/1Wx54si5_24rz0kJie31X54PIk_k_owT6qzlziGnAWYc/edit?usp=sharing). Next, clone this repository: +First, install the [Astrobee flight software](https://github.com/nasa/astrobee) in the '$HOME' directory. Additional instructions for using the Astrobee flight software with Docker are [here](https://docs.google.com/document/d/1Wx54si5_24rz0kJie31X54PIk_k_owT6qzlziGnAWYc/edit?usp=sharing). Next, create a catkin workspace and clone this repository into the catkin workspace. ```bash -cd $HOME && git clone https://github.com/hollydinkel/astrobee_change_detection --recurse-submodules +cd $HOME && mkdir -p change_ws/src && git clone https://github.com/hollydinkel/astrobee_change_detection --recurse-submodules ``` -Follow the steps in the provided `process_data.sh` file to download raw `.bag` data and process it into a change detection-compatible dataset. Note that the home directory, the change detection catkin workspac, the survey number (e.g., 1, 2, 3), the date (e.g., 20230419), and the robot name (e.g., bsharp) must be specified in the `process_data.sh` script. If the `astrobee` catkin workspace does not build locally, it is possible to do these steps in a docker container where the running container is mounted to a local directory. The provided `process_data.sh` script runs these four steps. +Follow the steps in the provided `process_data.sh` file to download raw `.bag` data and process it into a change detection-compatible dataset. Note that the home directory, the change detection catkin workspace, the survey number (e.g., 1, 2, 3), the date (e.g., 20230419), and the robot name (e.g., bsharp) must be specified in the `process_data.sh` script. If the `astrobee` catkin workspace does not build locally, it is possible to do these steps in a docker container where the running container is mounted to a local directory. The provided `process_data.sh` script runs these four steps. 1. The first step of data processing is extraction of images and poses from bag data to folder, processing of sequential images to remove images where frames did not move much between images, and processing of sequential poses so that pose timestamps are close in time to image timestamps. First, start a running Astrobee docker container with