ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
1 | initial version |
There is already a similar question question regarding this.
It can be a bit confusing, but according to the convention (it was maybe in REP 105) and the nav_msgs/Odometry message description, the twist should be in the base_link
. It actually makes sense to consider the twist locally. For example, lets say the robot is moving with 1m/s
forward. A twist, with linear x=1m/s
makes sense in this case regarding the base_link
and the robots current velocity. Otherwise if it was in the parent (odom
) frame then this simple movement would be represented with a combination of x
and y
.
2 | No.2 Revision |
There is already a similar question question regarding this.
base_link
. It actually makes sense to consider the twist locally. For example, lets say the robot is moving with 1m/s
forward. A twist, with linear x=1m/s
makes sense in this case regarding the base_link
and the robots current velocity. Otherwise if it was in the parent (odom
) frame then this simple movement would be represented with a combination of x
and y
. 3 | No.3 Revision |
There is already a similar question question regarding this.
It can be a bit confusing, but according to the convention (it was maybe in REP 105) and the nav_msgs/Odometry message description, the twist should be in the base_link
. It actually makes sense to consider the twist locally. For example, lets say the robot is moving with 1m/s
forward. A twist, with linear x=1m/s
makes sense in this case regarding the base_link
and the robots current velocity. Otherwise if it was in the parent (odom
) frame then this simple movement would be represented with a combination of x
and y
. regarding the robot's orientation in the odom
frame.