search

ethz-asl / mav_trajectory_generation

Polynomial trajectory generation and optimization, especially for rotary-wing MAVs.
Apache License 2.0
558 stars 225 forks source link

2nd constraints problem #68

Open schweinebauch opened 6 years ago

schweinebauch commented 6 years ago

Hello ,today I test tested the code and I added the 2th order constraints on one of the middle point.I found the trajectory got much more illogical than only position constraints added.About time allocation part I used 'runSegmentViolationScaling' function and just tested the linear optimization. I attached my trajectory and listed the coordinates of points. Could you help me solve this problem? What will affect the shape of trajectory generated? The reason is this optimization method is reformed as an unconstraints optimization? I do not need velocity constraints on way points? I am looking forward to receiving your feedback. Thanks in advanced! core_points (0, 0, 0)); //start point core_points(0.9, 1, 0)); core_points(0.7, 2, 0)); // at this point velocity constraints added 1m/s core_points(3, 3, 0)); core_points(4, 4, 0)); // goal point v_max = 2.0 a_max = 1.0 2018-07-17 16-35-11 @rikba @helenol

helenol commented 6 years ago

You generally don't need velocity constraints on waypoints; adding them overconstrains the problem, as you've seen (also keep in mind that the velocity is a 3-vector, so you need to set all the components of the velocity. Looks like it's set to point 1 m/s in the positive x direction, which is why the velocity looks so weird.