sangwoomoon/multiple_UAVs

how to use it ?

Opened this issue · 7 comments

danghc-bit commented
how to use it ?
sangwoomoon commented

Sorry for my late answer and the answer would not be enough since the code was about a decade ago.

Two folders represent essential approaches for operating potential function-based collision avoidance and Line-based path planning which has the MILP/NBTA-based task allocation. These two are hierarchically connected by feeding path planning results into the collision avoidance.

  1. Main m-files in Path planning and task allocation
  • Line_Algorithm_Cluttered.m
  • Line_Algorithm_Mapping_X_plane.m
  • Line_Algorithm_Unknown.m
  • Task_Assignment.m
  • simulation_TA.m
  1. Main m-files in Potential function-based collision avoidance
  • Fundamental_Mission_Planner.m
  • Motion_Planning.m
  • Motion_Planning_2.m
  • Multi_agent_Mission_Planner.m
  • Multi_agent_Mission_Planner_2.m
  • 👍1
koo-ec commented

Thanks for providing useful information here. I have tried Fundamental_Mission_Planner.m but it seems that "scan_pts" is missing.

Unrecognized function or variable 'scan_pts'.

Error in Fundamental_Mission_Planner (line 298)
plot3(scan_pts(:,1),-scan_pts(:,2),-scan_pts(:,3),'r.');"

koo-ec commented

Also for running "Line_Algorithm_Cluttered.m" I am getting the following error:

Unrecognized function or variable 'obstacle_input'.

Error in set_obstacle (line 15)
vrt((OBS_VRT+1)(idx_bldg-1)+idx_edge,1) = obstacle_input((OBS_VRT+1)(idx_bldg-1)+idx_edge,1);

Error in Line_Algorithm_Cluttered (line 102)
vrt = set_obstacle(OBS_NUM,OBS_VRT,0,1);

7keqi commented

Did you save the problems?

up-ls commented

I also encountered the same problem when running "Line_Algorithm_Cluttered.m". Did you solve the problems? @koo-ec

koo-ec commented

Hi @up-ls, @7keqi,

No, unfortunately, I could not fix the issue.

sangwoomoon commented

Hello @koo-ec, @up-ls, and @7keqi

So sorry for very late responses because I just found your updates right before. Hope it would help to fix the issues:

scan_pts variable generates when the mock laser scanner detects obstacles. Can you check if all obstacles are out of range with respect to the robot location? It does not look critical because the error happens from visualization, but please let me know if you always have errors in any conditions.

obstacle_input is obtained from Obstacles_geometry.m. But the problem is we need to load Obstacles_LLH.mat which cannot be served [link]. However, you can mimic the data by generating obstacle vertices information in longitude-latitude-height (LLH) coordinates. With my understanding, the format should be: matrix of which row is [vertex index, latitude, longitude] [link].