Adrian Egli
Adrian Egli
Excellent idea. I will think about the problem how we could provide a simple multi agent deadlock avoidance implementation to compare runtimes
The issue can be closed, right? mmarti-tsch ***@***.***> schrieb am Mi., 12. Juli 2023, 16:26: > I found out that this is not so much an issue of the reset()...
I have as well implemented a transformation from flatland (grid) into a graph representation: https://github.com/aiAdrian/flatland_railway_extension/blob/master/flatland_railway_extension/FlatlandGraphBuilder.py The FlatlandGraphBuilder converts Flatland's grid cell-based topology into a directed graph g. The graph consists...
What is the env? env.render i dont know the method
Would it be possible to test whether the RenderTool and Flatland could benefit from your renderToVideo experiments? See also: https://github.com/flatland-association/flatland-rl/blob/main/notebooks/simple_animation_demo.ipynb who to access raw rendered image from Flatland's RenderTool. See...
def make_simple_rail() -> Tuple[GridTransitionMap, np.array]: .... return rail, rail_map, optionals The return is not equal to the "typing" defintion. -
See also: https://github.com/aiAdrian/flatland_solver_policy/tree/main/observation/flatland/flatten_tree_observation_for_rail_env
@gdalle mentioned that rustworkx is a fast graph libary https://arxiv.org/abs/2110.15221 https://qiskit.org/ecosystem/rustworkx/dev/ https://github.com/Qiskit/rustworkx
I did a first implementation: https://github.com/aiAdrian/flatland_solver_policy
What is the current state of this PR? Still open?