search

proroklab / VectorizedMultiAgentSimulator

VMAS is a vectorized differentiable simulator designed for efficient Multi-Agent Reinforcement Learning benchmarking. It is comprised of a vectorized 2D physics engine written in PyTorch and a set of challenging multi-robot scenarios. Additional scenarios can be implemented through a simple and modular interface.
https://vmas.readthedocs.io
GNU General Public License v3.0
335 stars 69 forks source link

Lidar sensor's ray angles are fixed relative to world frame, but not to Agent frame #100

Closed wt160 closed 5 months ago

wt160 commented 5 months ago

In the current implementation, even the Agent's angle could be changed, but the sensor Lidar's ray angles are fixed relative to the world frame, which is not obvious when the lidar has 360 fov. Since the lidar supports different fovs, it seems the current configuration is kind of anti-intuition.

matteobettini commented 5 months ago

Thanks a lot for this, I'll fix it right away

matteobettini commented 5 months ago

Let me know if the PR works for you