Antonio Loquercio
Antonio Loquercio
MPC is our low-level controller, which we always use
I am very sorry that you are having trouble reproducing the work in the real world. Happy to help! Without knowing the details of your real-world system is difficult to...
We will release all hardware-related details plus a tutorial on building in the next few months. Keep an eye at our webpage!
Hi, here a codebase to bring this to the real world: https://github.com/uzh-rpg/agilicious
try ```sudo apt-get install libjpeg-dev zlib1g-dev```
Basically, we have a small warm-up stage (~1sec) to get to cruise speed. Eliminating the transient problem makes the learning problem easier. This requires to start with no obstacles in...
Hi! I believe that our work could definitely support vision-based swarm navigation. I would envision the following system: a centralized system giving high-level goals to each swarm member, and each...
You need to publish the depthmap as a rostopic. Just change the topic name in [here](https://github.com/uzh-rpg/agile_autonomy/blob/main/planner_learning/config/test_settings.yaml#L4) and you're done :)
Hi! We noticed similar behaviour also during experiments (large drift over the z axis). Our realsense does not perform perfectly at all, but it always provides a (somewhat consistent) estimate...
Rendering happens on GPU, while CPU controls the physics of the quadrotor