Open GongYanfu opened 2 years ago
I think it should be gmapping.
We do not construct occupancy maps I believe.
Do you sure that the map of the real environment is not needed for getting images ?
Positive. I collected the images by manually passing the sequence of positions.
From: GongYanfu @.> Sent: Tuesday, September 6, 2022 1:04:10 PM To: jkulhanek/robot-visual-navigation @.> Cc: Jonáš Kulhánek @.>; Comment @.> Subject: Re: [jkulhanek/robot-visual-navigation] puzzle about the construction of occupancy grid map (Issue #18)
Do you sure that the map of the real environment is not needed for getting images ?
— Reply to this email directly, view it on GitHubhttps://github.com/jkulhanek/robot-visual-navigation/issues/18#issuecomment-1238054362, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABAGZIVTYYWC3IB2XYAJQITV44XLVANCNFSM54RZRMJA. You are receiving this because you commented.Message ID: @.***>
In the virtual environment, how does the algorithm determine whether the agent has reached the goal? I didn't find the corresponding code.
This function is_goal is for the environment constructed by images we got from real environment, no for the virtual environment.
Oh, I see, I am sorry, the code is here: https://github.com/jkulhanek/dmhouse/blob/c68e9a4f0f4ac5c5e73ea96d7c9b8fcdaa3a68d9/game_scripts/custom/hit_goal_decorator.lua#L21
What algorithm is used for the occupancy grid map in the paper? gmapping, hector-slam or else? perhaps this is not the key point, but it puzzles me.