The site is read-only. Please transition to use Robotics Stack Exchange
| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
 | 1 | initial version |
I figured another way out. I passed Vector3 to my function instead of Float64 and it worked. But still a way for type conversion might be helpful.
There is no conversion defined for Vector3 to Float64 (or really double, as the error message tells you). I'm also dubious as to how that would work: C++ type conversions are from one value to another, while Vector3 packs three values, so should return another container-like type (a tuple, vector or something like that) in order to work.
Is there anyway I can pass vector3 values as vector3.x as parameters to my function ?
Well, yes: geometry_msgs/Vector3 contains x, y and z members, didn't the following work for you?
calc_cartezian_trajectory(..., vector3.x, vector3.y, vector3.z, ...)
If that is really a problem, you could perhaps create an overloaded version of calc_cartezian_trajectory(..) that does take a geometry_msgs::Vector3 instance, and then calls the one you have now.
PS: slightly off-topic, but cartezian -> cartesian.
| | 2 | No.2 Revision |
I figured another way out. I passed Vector3 to my function instead of Float64 and it worked. But still a way for type conversion might be helpful.
There is no conversion defined for Vector3 to Float64 (or really double, as the error message tells you). I'm also dubious as to how that would work: C++ type conversions are from one value to another, while Vector3 packs three values, so a conversion should return another container-like type (a tuple, vector or something like that) in order to work.work. That wouldn't really help that much I believe.
Is there anyway I can pass vector3 values as vector3.x as parameters to my function ?
Well, yes: geometry_msgs/Vector3 contains x, y and z members, didn't the following work for you?
calc_cartezian_trajectory(..., vector3.x, vector3.y, vector3.z, ...)
If that is really a problem, you could perhaps create an overloaded version of calc_cartezian_trajectory(..) that does take a geometry_msgs::Vector3 instance, and then calls the one you have now.
PS: slightly off-topic, but cartezian -> cartesian.
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.