| ROS Resources: ROS Homepage | Media and Trademarks | Documentation | ROS Index | How to Get Help | Q&A Help Site | Discussion Forum | Service Status |
Hello folks,
Nemesis is a UVG project I worked on to explore the challenges associated with integrating an existing ground vehicle with the ROS ecosystem. https://antoan.github.io/Nemesis/
My main focus was the hardware interface implementation (adapted from the Husky Clearpath stack) and the perception stack I added later, not the robot build as much.
Cheers,
Antoan
Hello Neuromorphic i’m working on a big UGV which will be used in both indoor mapping and outdoor mapping and also will perform localization and path planning (ROS) can you be of gaidiance from you . I hope you dont mind guiding me out
Hi Daviesss,
Sure, I can try to point you in the right direction. Drop me an email describing the specifics of your project and I’ll see if I can help out: antoan.bekele@gmail.com
Thanks for your response Neuromorphic. I sent in an email describing how and what the project I’m working on is about.
I made a quite similar setup with RPi4 , d435 and t265
I will need comeback to it for some rework .
RIP t265
RIP indeed, such a shame. Why did they have to kill it.
Nice setup, btw. Do you have encoders on your wheels?
No, 360 2d lidar (ICP) and t254 is enough good, but I didn’t test extensively, it is 2 years old now, I didn’t have time to come back to it.
Well , technically t265 is a failed product, as d455 is. But people needed buy it and use it.No many alternatives, and that didn’t change.
I want to remove VIO and depth cameras of my life, at least as main component, just thinking in lidar , event cameras and whatever the AI can help.
At the moment I got a 3d lidar.
T265 is not reliable enough, but is ok for many cases and it was well priced.
Just a hobbyist personal opinion, and to be honest ,if you cannot get a lidar or event camera, I wouldn’t look to any other stereocamera of the market, the best alternative to t265, is buy the t261 module and the USB interface (difficult that)
I would not look other recent stereo cameras, but again just personal opinion from a hobbyist.
Enjoy your cool project
Thanks for your informative reply, buddy.
I wish I could get my hands on an event camera sensor.
It seems key to really agile/fast navigation that I am very interested in, due to the low microsecond latency, ideal for control. Are they not ridiculously expensive at the moment? I haven’t looked.
Something you might be interested in if you plan to do your own custom SLAM setup, that I plan to use my self, is symforce from the Skydio (Frank Dellawert author of GTSAM) drone team. It allows one to express a custom sensor configuration in python as a factor graph and generates optimised C++ code.
