kpm的文章看过后,照例跑跑代码,细节如下
We keep the implementations of all multi-robot motion planning algorithms in a library called deconflictiontools, see:
http://jones.felk.cvut.cz/redmine/projects/deconflictiontools/wiki/Wiki
The source code of this library can be downloaded through mercurial:
The source code of this library can be downloaded through mercurial:
hg clone http://jones.felk.cvut.cz/repo/alite-deconflictiontools
The implementation of kPM is in that library, specifically you may want to start in class tt.jointeuclid2ni.solver.impl.AlgorithmDPMD, which is an implementation of kPM that uses a graph-based motion planner to plan for each robot at each step.
mercurial是一个工具,用sudo apt-get install mercurial
安装
在solvers里,只用于scenarios里的两个例子
进入kpm-experiments-master\solvers
输入
mvn package
生成
target/kpm-experiments-1.0-SNAPSHOT-jar-with-dependencies.jar
输入
java -jar solvers/target/kpm-experiments-1.0-SNAPSHOT-jar-with-dependencies.jar -method KDPMD -k 20 -problemfile scenarios/headson.xml -maxtime 21000 -timestep 21 -showvis -summary
或者(不同xml文件)
java -jar solvers/target/kpm-experiments-1.0-SNAPSHOT-jar-with-dependencies.jar -method KDPMD -k 20 -problemfile scenarios/swap-in-corridor.xml -maxtime 21000 -timestep 21 -showvis -summary