===== Task10 - Data collection path planning with curvature-constrained trajectory - Dubins TSP with Neighborhoods (DTSPN) - decoupled approach   ======

|**Deadline**  |  15. December 2018, 23:59 PST |
|**Points**  |  3 |
|**Label in BRUTE**  |  Task10 |
|**Files to submit**  |  archive with ''DTSPNSolver.py'' |
|**Resources**  | {{ :courses:b4m36uir:hw:task10-v2.zip | Task10 sources (version 2) }} |

==== Installation of the prepared codes ====

Download prepared codes and configuration files. Then download and compile LKH solver (implementation of the Lin–Kernighan heuristic algorithm) as follows:

<code bash>
cd lkh
./install.sh
cd -

cd gdip
./install.sh
cd -
</code>

Now, you should be able to run the prepared examples.

<code bash>
./task10_eval.py
./gdip_example.py
</code>

==== Dubins maneuver ====
[[https://en.wikipedia.org/wiki/Dubins_path]]

==== Generalized Dubins Interval Problem ====
[[https://github.com/comrob/gdip]]

====  Tasks (3 points) - Task10 ==== 
  * Implement decoupled-based solution for the Dubins TSP with Neighborhoods (DTSPN) with disk-shaped regions.
    * First, estimate sequence of visits by Euclidean TSP (ETSP) connecting centers of the regions.
    * Sample the boundary of the neighborhood and heading angle.
    * Find the shortest feasible tour created from Dubins maneuvers based on the prepared samples. Utilize the sequence estimated from the ETSP.

Notice, the number of heading and position samples is not defined. Your task is also to try various values and submit a code with a reasonable number of samples.