search

mikh3x4 / nerf-navigation

Code for the Nerf Navigation Paper. Implements a trajectory optimiser and state estimator which use NeRFs as an environment representation
https://mikh3x4.github.io/nerf-navigation/
MIT License
186 stars 24 forks source link

a star init problem when running simulate.py #11

Closed RONINGOD closed 1 year ago

RONINGOD commented 1 year ago

(Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2228.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] (13, 3, 12) (6, 15, 11) Traceback (most recent call last): File "simulate.py", line 353, in simulate(planner_cfg, agent_cfg, filter_cfg, extra_cfg, density_fn, render_fn, get_rays_fn) File "simulate.py", line 46, in simulate traj.a_star_init() File "/nerf-navigation/nav/quad_plot.py", line 95, in a_star_init path = astar(occupied, start, end) File "/nerf-navigation/nav/quad_helpers.py", line 219, in astar assert not occupied[start] AssertionError

Prasaya commented 1 year ago

@RONINGOD Did you solve this?

RONINGOD commented 1 year ago

@Prasaya Yes

Prasaya commented 1 year ago

@RONINGOD How did you solve this? Which dataset did you use?

RONINGOD commented 1 year ago

@Prasaya I upgrade PyTorch version to 1.12.0 and use origin stonehenge dataset.