Only released in EOL distros:
Package Summary
The nj_oa_costmap package is a navigating jockey for the Large Maps framework. Its role is to drive the robot forward while avoiding obstacles. It uses a local map which position is relative to the robot but which orientation is constant.
- Maintainer status: developed
- Maintainer: Gaël Ecorchard <gael.ecorchard AT ciirc.cvut DOT cz>
- Author: Gaël Ecorchard <gael.ecorchard AT ciirc.cvut DOT cz>, Karel Košnar <kosnar AT labe.felk.cvut DOT cz>
- License: BSD
- Bug / feature tracker: https://github.com/lama-imr/lama_costmap/issues
- Source: git https://github.com/lama-imr/lama_costmap.git (branch: indigo-devel)
The nj_oa_costmap package implements a navigating jockey for the Large Maps Framework (LaMa) that drives the robot while avoiding obstacles. It is based on a local map (nav_msgs/OccupancyGrid). The local map has a fixed position relative to the robot but its orientation is constant in the world reference frame. It is a reactive, memory-less jockey.
Implemented action:
- TRAVERSE: go more or less forward depending on obstacle position. The action never stops by itself.
- STOP: stop the robot, sending this action is the only way to stop a TRAVERSE action.
- INTERRUPT: same as STOP
- CONTINUE: same as TRAVERSE
Moreover, the package provides the class TwistHandler that computes an appropriate geometry_msgs/Twist from a nav_msgs/OccupancyGrid in order to go as forward as possible while avoiding obstacles.
Usage
ROS API
Subscribed Topics
~<name>/local_map (nav_msgs/OccupancyGrid)- local map with fixed position relative to the robot and fixed orientation relative to the world reference frame
Published Topics
~<name>/cmd_vel (geometry_msgs/Twist)- set velocity
Parameters
~<name>/robot_radius (Float)- robot radius (m)
- if an obstacle is closer than this, turn and don't go forward (m)
- if no obstacle within this distance, go straight (m)
- turn rate when obstacle closer than ~min_distance_ (rad/s)
- linear velocity without obstacle (m/s)
- linear velocity if obstacle between min_distance and long_distance (m/s)
- if obstacle closer than ~long_distance, turnrate = -turnrate_factor_ * mean(lateral_position_of_obstacle) (rad.m-1.s-1).
- frame_id of the laser scan messages that are used to build the local map.
- Jockey name