Skip to content

Latest commit

 

History

History
40 lines (26 loc) · 1.84 KB

Readme.md

File metadata and controls

40 lines (26 loc) · 1.84 KB

PANTHER: Perception-Aware Trajectory Planner in Dynamic Environments

Main repo is this one

This repo simply contains files that are needed either on the base station (for takeoff/land high-level commands) and on the SnapDragon.

This repo should be cloned as a submodule of PANTHER on the Intel NUC, and as a standalone repo on the base station and SnapDragon

Notes to run HW experiments:

The param used in these files

	predictor.yaml
	panther.yaml
	panther/matlab/casadi_generated_files/params_casadi_prediction.yaml  [To change the parameters of this file, you must change it in predictor.m, and run that file again]
	panther/matlab/casadi_generated_files/params_casadi.yaml  [To change the parameters of this file, you must change it in main.m, and run that file again]

are available in tag RAL_hw (commit 922e2c1d5bcb693ae of PANTHER). But check also these changes. Additionally, you may want to change (for better performance) linear_solver_name='mumps' to linear_solver_name='ma57' (or ma27) in main.m

To install sympy in the SnapDragon or Sykorsky:

sudo apt-get install python-sympy --no-install-recommends

Other

In Sykorsky: roslaunch behavior_selector gui.launch

Trajs of the obstacles

The file generate_traj_params.py will generate the .yaml files that specify the trajectories of the obstacles. These files are then used by obstacle.launch, which creates the corresponding parameters needed in obstacles.py.

Approval for release: This code was approved for release by The Boeing Company in March 2021.