Embedded Motion Control 2015 Group 3/Experiments

From Control Systems Technology Group
Jump to navigation Jump to search

Experiments

This page is part of the EMC03 CST-wiki.

Experiment 1

Date: Friday May 8

Purpose:

  • Working with PICO
  • Some calibration to check odometry and LRF data

Evaluation:

  • There were problems between the laptop we used and PICO.

Experiment 2

Date: Tuesday May 12

Purpose:

  • Calibration
  • Do corridor challenge

Evaluation:

  • Here the results of the calibration can be found.
  • Driving through a straight corridor went very well. But we could not succeed the corridor challenge yet.

Experiment 3

Date: Tuesday May 22

Purpose:

  • Calibration
  • Do corridor challenge

Evaluation:

  • Combining the Scan and Drive for path planning was not succesfull.
  • Potential field script was not ready yet.

Experiment 4

Date: Tuesday May 29

Purpose:

  • Test potential field
  • Test path planning

Evaluation:

  • Path planning was not very succesfull.
  • Potential field did very well in corridors (see video). Intersections need some extra attention.

Experiment 5

Date: Friday June 05

Purpose:

  • Test potential field & decision intergration

Evaluation:

  • Integration is almost done. Still has some problems with crossing intersection and taking corners into (short) dead ends like doors.
  • The collision detection behaves somewhat unexpected some times.

Experiment 6

Date: Friday June 12

Purpose: Test everything combined. Especially, some difficult situation with doors, open spaces, dead ends and loops.

Evaluation:

  • Localisation was not finished yet.
  • There were some problems with the angular corrections. Therefore, PICO sometimes drove sideways through the corridor.
  • We have make adjustments before next experiment on Tuesday June 16.

Experiment 7

Date: Tuesday June 16

Purpose:

  • The last repetition for the A-maze-ing challenge

Evaluation:

  • Sometimes PICO made strange choices at intersections. Probably, because the localisation was not working well.
  • Therefore, before tomorrow the localisation has to be adjusted. And we will make a back-up plan that is random and uses no localisation at all.