Goose Chasing Robot

Hi,

I have a terrible problem with geese in my yard. I’ve also been a heavy ROS user for the past 2 years now so I thought I’d make a robot to help chase the geese away. I have a prototype up and running, it can visually recognize geese up to 20ft away and will navigate to there location at 0.5m/s. I’d like some help on a few things if anyone is interested.

Things I want:
A better state machine
Autonomous docking
Mobile app support (I think this looks like a good project to work off of https://github.com/mtbii/RobotCA)

The first thing I could use help on is the state machine. If anyone has experience with using ros_smach as is willing to help (or just advise), please email me (nicholas.fragale@gmail.com)

4 Likes

Hey @shoemakerlevy9, cool project :slight_smile:

In terms of the state machine, here are some resources that might be of interest:

1 Like

For mobile support, I don’t know how deep you want to get into mobile development but you could look into rosjav/android: http://wiki.ros.org/android

@jcerruti
That would have ROS running on an Android device right? I think I’d rather have the master running on a Linux board and just have a tablet for veiwing information. I have a NVidia Jetston TX1 that I’m running my master on now. That has enough power to run a neural net for recognizing geese.

@spmaniato

Great! This is exactly the type of info I was looking for. Thanks you!

P.S.
I also went to Cornell. I’m the one who emailed you about NASA SRC advice :slight_smile:

@shoemakerlevy9 you can have your master running in the linux board, and an Android tablet using rosjava to connect to that master, subscribe to topics and view information from there. Creating a client application using ROS Android / ROS Java is not that hard, and it may serve your purpose very well!

BTW, cool project! I would love to see a video of the robot chasing geese.

You may find some inspiration in Richard Vaughan’s PhD work: https://youtu.be/tefXVXscNDM :slight_smile: There is a link in the video description with more details.

2 Likes

@bgromov

Wow, that will be immensely helpful :slight_smile: Thank you!

@shoemakerlevy9 @bgromov Here’s the paper [Vaughan, RAS 2000]. Enjoy!

2 Likes