Search found 323 matches

by whoenig
Sun Feb 23, 2020 4:49 am
Forum: Support
Topic: Accuracy of trajectory calculated on 1-D
Replies: 1
Views: 1688

Re: Accuracy of trajectory calculated on 1-D

I usually use the logging variables "stateEstimate.{x,y,z}" and "ctrlTarget.{x,y,z}". The state estimate combines your mocap measurement with the oboard IMU. The best way is to log those variables using the uSD-card deck, and then use post-processing to analyze the data. However, you can also stream...
by whoenig
Sun Feb 23, 2020 4:44 am
Forum: Autonomous flight
Topic: How to generate trajectory
Replies: 3
Views: 2589

Re: How to generate trajectory

You can follow the example at https://github.com/whoenig/uav_trajecto ... tion-pairs to generate your trajectory (csv-file). This file can be uploaded to the CF before the flight.
by whoenig
Fri Feb 21, 2020 6:08 am
Forum: Bitcraze
Topic: Optitrack to Crazyflie Set up
Replies: 8
Views: 1144

Re: Optitrack to Crazyflie Set up

Yes. Each PC just goes to the LAN port. You can also connect to the Internet by connecting the router's Internet port to your campus network, but that is optional.
by whoenig
Tue Feb 18, 2020 12:44 am
Forum: Bitcraze
Topic: Optitrack to Crazyflie Set up
Replies: 8
Views: 1144

Re: Optitrack to Crazyflie Set up

Direct Ethernet connection: Yes, no need for a DHCP server. You can configure a static IP in your OS network settings, which is OS dependent (you should be able to find instructions using a search engine). A router is a bit more convenient (plug&play). NatNetSDK: No, the crazyswarm comes with the SD...
by whoenig
Mon Feb 17, 2020 7:42 am
Forum: Bitcraze
Topic: Optitrack to Crazyflie Set up
Replies: 8
Views: 1144

Re: Optitrack to Crazyflie Set up

The Optitrack documentation should have some more information. For most mocap systems it is recommended to use multiple network cards on the Windows PC, one for Optitrack and one to connect to other computers. You can either directly connect the two PC using ethernet (and define static IPs), or more...
by whoenig
Sun Jan 05, 2020 11:22 pm
Forum: Support
Topic: Library of crazyflie in ROS documents
Replies: 2
Views: 1042

Re: Library of crazyflie in ROS documents

This refers to the helper library in https://github.com/whoenig/crazyflie_ro ... azyflie.py. It should be found automatically, as long as you run python with crazyflie_ros/crazyflie_demo/scripts as your current working directory.
by whoenig
Tue Dec 24, 2019 8:31 pm
Forum: Bitcraze
Topic: Try to figure out the ROS package
Replies: 2
Views: 1073

Re: Try to figure out the ROS package

Hi Duncan, 1. Some sort of localization in the demo scripts is required. However, this can be the flowdeck, LPS, Lighthouse, or mocap. In your case, you would essentially replace the feedback from such systems with the SLAM code. I suggest using the high-level-mode, see https://github.com/whoenig/cr...
by whoenig
Sun Dec 01, 2019 8:17 pm
Forum: General discussions
Topic: More powerful motors
Replies: 2
Views: 1105

Re: More powerful motors

I used https://www.amazon.com/BETAFPV-19000KV-Brushed-JST-1-25-Connector/dp/B07HD854SG/ref=sr_1_3?keywords=betafpv+7x16mm&qid=1566084299&s=gateway&sr=8-3 with the stock Crazyflie propellers before. The motors are much more powerful (~2.6 thrust-to-weight ratio rather than 1.5), but they also tend to...
by whoenig
Thu Nov 28, 2019 1:30 am
Forum: General discussions
Topic: Crazyflie 2.0 IMU data
Replies: 4
Views: 2180

Re: Crazyflie 2.0 IMU data

It depends on what variables you log. In this case it's just non-SI units to SI units, see https://github.com/whoenig/crazyflie_ro ... #L596-L604.
by whoenig
Sun Nov 24, 2019 4:17 am
Forum: Loco Positioning System
Topic: Script to compress trajectories
Replies: 4
Views: 1066

Re: Script to compress trajectories

Perhaps https://github.com/whoenig/uav_trajectories#generate-trajectory-given-time-position-pairs helps: You can encode a list of timed waypoints to trajectories with a specified number of pieces, i.e., if you want to reduce the memory you can reduce the number of pieces. The script solves an optimi...