eband_local_planner

The eband_local_planner is a ROS package, implementing a plugin to the base_local_planner for the move_base 2D navigation system.

It implements the Elastic Band method on the SE2 manifold.

Currently support indigo, kinetic distro, and melodic under testing.

Changelog for package eband_local_planner

0.4.0 (2020-02-14)

• Adds Dynamic Reconfigure for planner parameters
• Updates library usage from tf to tf2 as necessary for Melodic
• Release for ROS Melodic (closes #28 )
• Contributors: Maxwell Svetlik, Yifeng Zhu

0.3.1 (2017-07-02)

• Find Eigen3 and not Eigen to compile cleanly on Kinetic (#21)
• For Indigo, find Eigen if Eigen3 not found (#21)
• Contributors: Jack O'Quin

0.3.0 (2015-03-31)

• slow down trajectory controller while close to goal. closes #19.
• convert info message to debug. closes #18.
• clear local costmap if global plan cannot be converted to band even on first attempt. see #5
• Contributors: Piyush Khandelwal

0.2.2 (2015-03-14)

• fixed catkin lint warnings in preparation for v0.2.2 release
• added a bit more paramterization of the algorithm to help navigation in simulation. all changes are backwards compatible. #17 from utexas-bwi/piyushk/fix_simulation_navigation
• made some changes for a future Indigo release. fixed Eigen dependency (#16)
• Contributors: Jack O'Quin, Piyush Khandelwal

0.2.1 (2014-04-25)

• added jack as a maintainer for the package
• 0.2.0
• updated changelogs in preparation of v0.2.0
• changing some fudge factors
• removed band hack. attempting band repair algorithm now
• changes to hysteresis loop. still has some problems
• fixed indentation. closes #10
• eband_local_planner now fails when new frames cannot be added to the band. closes #9
• Contributors: Piyush Khandelwal

0.1.2 (2013-08-12)

• stopped publishing empty bubble heading message. closes #6
• reenabled holonomic drive control. still untested since groovy/hydro migration. closes #2

0.1.0 (2013-07-10)

• add include directory to installation
• removed redundant file
• introduced some hysteresis between turning to final orientation and turning towards the next bubble. closes #3
• returning trajectory controller's opinion on whether the goal has been reached or not. closes #4
• removed unnecesary print statements
• some debug statements to help find issues with hydro-devel costmap
• fixed all runtime loading errors in process of catkinization
• removing unneeded/untested launch files. I'll add working examples later as the package sees a public release.
• catkinizing eband_local_planner against the hydro-devel branch of navigation
• removing standalone follow trajectory ndoe
• promoting eband local planner to separate repository in preparation for catkinization
• some improvements to navigation
• fixed a bug in the eband trajectory controller
• Merge branch 'master' of github.com:utexas-bwi/segbot_apps
• in-place rotation at goal now supported
• merged goal tolerance parameters between local planner and trajectory controller.
• add launch for e-band navigation
• changed distance check to not include orienation. I am not even sure why orientation was being used to check overlap.
• a couple more parameters + better velocity limit checking
• fixed for the regular nav stack launch file as well. closes #1
• hmm not sure why this file was here
• fix for the eband costmap having an incorrect topic. #1
• updated launch file to use any visualization configuration + reorganized eband configuration file
• fixed a problem with requesting obstacle distances at the edge of the costmap - these regions are now assumed to be free (optimism) - removes problem with the global plan being regenerated over and over again as new parts of the global plan were converted to the band
• hack for global planner not catching up in case of an unexpected obstacle snapping band
• changed INFO statements to DEBUG
• checking in new parameters for the eband local planner
• inital differential drive trajectory controller - looks pretty good. needs a bit more code improvent, dynamic reconfigure and stricter obstacle testing
• Revert "checking in Paul's changes to get eband local planner to work with our differential drive robot". I am seeing some problems with his changes, will apply by hand. This reverts commit c925f7e1d5bebbf8c102a2292429749da84562a9.
• checking in Paul's changes to get eband local planner to work with our differential drive robot
• added git ignore settings
• changes to make eband_local_planner compile with navigation 1.9 + groovy
• copy of eband_local_planner from navigation_experimental.
• removed some unnecessary launch files and added an rviz configuration + launch file for testing autonomous navigation
• basic amcl + move base demo works (but is not very good)
• removed old ens basement maps from the repo
• removed joy gmapping file - joystick control not directly supported
• removed redundant sensor files (moved to segbot_sensors)
• initial port of of navigation and controller code from the svn repository
• first commit