https://github.com/varun0157/image-based-visual-servoing

Image Based Visual Servoing (IBVS) simulation in pybullet
https://github.com/varun0157/image-based-visual-servoing

ibvs robotics robotics-simulation visual-servoing

Last synced: 2 days ago
JSON representation

Image Based Visual Servoing (IBVS) simulation in pybullet

• Host: GitHub
• URL: https://github.com/varun0157/image-based-visual-servoing
• Owner: Varun0157
• License: mit
• Created: 2024-04-05T05:52:38.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-11-08T08:00:07.000Z (about 1 month ago)
• Last Synced: 2024-11-08T09:17:22.569Z (about 1 month ago)
• Topics: ibvs, robotics, robotics-simulation, visual-servoing
• Language: Python
• Homepage:
• Size: 26.7 MB
• Stars: 4
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
### Instructions

- `src/env.yml` contains the conda environment config file. Create an environment (here, called `ibvs`) and load it in as usual:

```bash 

cd src 

conda env create -f env.yml

conda activate ibvs  

```

- run the pybullet sim in `DIRECT` mode, with images saved to `src/img/`. 

```bash

cd src 

python index.py

```

- if required, convert the images into an MP4 file for later analysis, storing the video in `src/vid/`

```

bash create_vid.sh

```

#### Demo 

|  |  |

|:---:|:---:|

| Target Image | Output GIF |

### TODO

- [ ] port to `airobot` to get rid of view matrix complexity

- [ ] use a better target and scene, add some acknowledgements 

- [ ] experiment with more complex dof. Currently simply taking the first three points' (x, y) coords as our 6 dof. 

- [ ] find out why the obstacles fall slowly in the p.DIRECT sim even though it does not seem to happen in p.GUI.