juanlu's profile - activity

Having same problem on ROS Melodic and pcl_ros 1.8.1

Hi,

Im implementing an explorer algorithm that takes the map from a SLAM algorithm, processes it and sends a goal to move base. The problem is that sometimes the goal is not away enough from a wall and move base takes a long time to stop replanning and reject the goal (up to several minutes).

I saw the parameter "default_tolerance", by planning to a goal within a tolerance raidus, should be the answer to my prayers but a bit of investigating here in ros answers and in the code showed me that the parameter has no effect whatsoever in the execution of the code.

Does anybody have any suggestion on how to fix this?

Thanks in advance.

How did you solve it?

At first sight I think the fixed frame is not correct, can you try setting it up to base_link?

Hi, the data doesn't look bad to me, I'm assuming the "I" represents Infinite for the lidar rays that aren't hitting any

"The package subscribes to cmd_vel and publishes odometry based on that" From the wiki of diff drive controller: Od

I use TEB daily and there is no jerking whatsoever, it should be possible to achieve what you want using it. Maybe check

This will get you the center of the grid cell with "map" being a nav_msgs/OccupancyGrid: map_x_coordinate_in_meters = (

This will get you the center of the grid cel with "map" being a nav_msgs/OccupancyGrid: map_x_coordinate_in_meters = (g

This will get you the center of the grid cel with "map" being a nav_msgs/OccupancyGridl: map_x_coordinate_in_meters = (

This will get you the center of the grid cell: map_x_coordinate_in_meters = (gridcell_x_coordinate * map.info.resolutio

This will get you the center of the grid cell: map_x_coordinate_in_meters = (gridcell_x_coordinate * map.info.resolutio

This will get you the center of the grid cell: map_x_coordinate_in_meters = (gridcell_x_coordinate * map.info.resolutio

This will get you the center of the grid cell: map_x_coordinate_in_meters = (gridcell_x_coordinate * map.info.resolutio

The goal of the global planner is to find the shortest path between your current position and the goal, It does so by us

The feature you want is called "Lifelong mapping" and as far as I know there is no available SLAM algorithms in ROS that

The feature you want is calle "Lifelong mapping" and afaik there is no available SLAM algorithms in ROS that offer it.

The metadata of the maps contains a field to indicate the resolution. If you open a map in rviz and look at its paramete

To compile your code in release you have to add this flag to the CMakeLists, taken from here: if(NOT CMAKE_CONFIGURATIO

To compile your code in release you have to add this flag to the CMakeLists, taken from here: if(NOT CMAKE_CONFIGURATIO

Since you don't have encoders and will have to rely completely on the camera to determine the future position of the obj

Since you don't have encoders and will have to rely completely on the camera to determine the future position of the obj

I see, then I am out of ideas. Try debugging the arduino while connected to a computer using the monitor ,checking that

In your code I am missing the serial part from the PC, how are you sending those msgs to the arduino? I also see in the

To compile your code in release you have to add this flag to the CMakeLists: set(CMAKE_BUILD_TYPE Release) Finding ou

You are right! Ill edit my answer

Hi! before answering let me point you to the robot localization wiki, which is great and has a lot of information and t

To compile your code in release you have to add this flag to the CMakeLists: set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_F

Do you export the wheels separately? Check the urdf to see if the wheel link and joint are present and correct, compare

Are the "base_parameters" correctly calibrated? I found this two parameters in a launch file in the repo you posted: &l

Question, do you have a full robot or just a single DC motor?

Great answer! I'd like to add that in case he is not interested in writing his own controllers and odometry computation

Yes, then I am afraid you have to implement an additional outlier rejection method to discard those values with high cov

Yes, then I am afraid you have to implement an additional outlier rejection method to discard those vales with high cova

Where is the rate 1Hz, from PC to Arduino or from Arduino to steppers? Also, would you have any idea what could be givin

Hi,

I am trying to enable a Mecanum Wheeled Robot in ROS Control and I am having a lot of trouble understanding what it is that I have to do.

From what I understand the data flow is as such:

I'm using the Ridgeback mecanum controller and through arduino I have access to the wheels velocities. I have read the ros control documentation and seen the 2014 ROSCon video about ROS Control but I'm still missing that last piece to understand it.

Cheers!

So I actually achieved understanding of this issue and my poorly formulated question, here it is: My actual question wa

Sorry, stupid question. Could it be that your terminal has a character limit that doesn't let you scroll? Anyhow, maybe