Piecewise Polynomial Tracking Question

[Iko-Git]

Hello!
Could you please guide me or refer me to a clear source whereby I can learn how to execute and simulate 7th degree piecewise polynomials on multi CFs starting from a certain waypoint in the space for every CF. Providing an example would be highly appreciated

It is important to mention that all my agent's trajectories have the same shape of figure8.csv example

[whoenig]

I am not entirely sure what you are looking for. Perhaps slides 16 to 18 in https://drive.google.com/file/d/15favAyrLLpC_O6nrAp-eIbZijFUMLgwV/view helps.

[jpreiss]

That will be straightforward to implement. You can start with swap6v.py and make a few modifications for your scenario.

[Iko-Git]

Thanks for your comment. I am generating a trajectory on Matlab. The trajectory is generated as .csv file which has the same structure as the .csvs used in swap6v.py example. The problem is that the trajectory plot I am seeing on Matlab is different than the CF behavior in the simulator.
Below you can find an example on the csv I am using, and the plot of the trajectory.

agent1.csv

[whoenig]

You can plot your csv file using https://github.com/whoenig/uav_trajectories/blob/master/scripts/plot_trajectory.py. Make sure that this looks like you expect, otherwise you might have used a slightly different encoding (order of polynomial coefficients?).

[Iko-Git]

Thanks for your remark. looks to be there was a mistake in the order of the coefficients!

Excuse me if the following two questions should not be posted under this discussion and please let me know if it has to be discussed in uav_trajectories repo.

1. In https://github.com/whoenig/uav_trajectories, trajectories can be generated based on a set of waypoints with constraints on the velocity and acceleration. E.g. ./genTrajectory -i examples/waypoints1.csv --v_max 1.0 --a_max 1.0 -o traj1.csv . Is there any resource I can be referred to in order to study how the generated optimized trajectories consider the velocity and acceleration constraints?
2. Are the trajectories generated by the mentioned repo snap optimized?

Thanks again for your support!

[whoenig]

The repo is based on https://github.com/ethz-asl/mav_trajectory_generation, which in turn implements: C. Richter, A. Bry, and N. Roy, “Polynomial trajectory planning for aggressive quadrotor flight in dense indoor environments,” in International Journal of Robotics Research, Springer, 2016.