fmarques's profile - activity

JadTawil, I've seen in your other posts that you have a nanotec controller (PD6-C). Did you managed to make it work with

Hi, After reading http://answers.ros.org/question/9888/switching-global-planner/

I've got some questions regarding the issue of switching global planner. As I can understand the global planner is a parameter of move_base, and it's displayed in the dynamic reconfigure but I cannot change it online?

I need to use different planners and different configurations controlled by a supervisor node. For now the options are:

Kill the move_base node and relaunch it ( roslaunch api ???)

Keep two move_base nodes online and simply reroute the goals.

Somehow change all parameters on the fly using dynamic_reconfigure.

Anyone has experience and doing something similiar, I think this would be a great addition to ros Navigation.

Thanks

Hi, I'm using the sbpl_lattice planner and I have a question about how to generate my own motion primitives. My robot cannot perform foward motion arc that have a radius below 1 meter.

So I'm trying to modify the genprim_unicycle.

%1/16 theta change basemprimendpts0_c(4,:) = [4 1 1 forwardandturncostmult];

From what I can understand this is the X,y, theta displacement.

But I'm having problems in making smooth trajectories.

Does anybody already has experience in making your own motion primitives??

I've already altered the base_local_planner so it to follows the plan very accurately.

Thanks

Hi,

I'm working in a robot and started to use the hector mapping package in replacement of gmapping. I only have one doubt regarding the setting up your robot tutorial.

http://www.ros.org/wiki/hector_slam/Tutorials/SettingUpForYourRobot

How can I provide the transformation between the base_link and base_stabilized, i have a mems imu and robot pose ekf implemented. So I simply add a new frame /base_stabilized) to tf??? Can anybody that already has this working in their robot help me out... Thanks

Hi, I'm trying to use the laser scan matcher package in my mobile robot. So I've performed numerous simulations on gazebo and everything worked great.

Now i'm performing the first tests with the physical robot, and a problem occurred, at high speeds (linear or angular) i't gives very good results but when traveling at low speeds it tends to drift allot, is this a issue that is to be expected because of the signal to noise ratio at slow speeds or my parameters wrong (mostly default). The environment is a normal office with the particularity that two office cabinets don't reflect the laser but other than that it's standard and with plenty of features for the icp . Tomorrow I gonna perform tests in a more controlled environment (less cluttered) so I can rule that out. If anybody had or has a similiar problem please let me know.

Thanks

Hi,

I've asked a related question when I needed to put gmapping, laser_scan_matcher and stageros working together. The problem was that but stageros and laser_scan_matcher publish the odom->base_footprint. I solved the problem by going to the stageros.cpp and stop the publish of the transform.

But know I want to make a more evolved simulation including IMU and Cameras and wanted to try Gazebo, but the same problem occurs and I cannot find where modified Gazebo to stop publishing.

Has anyone had the same problem???

Thanks in advance

P.S. I think will look in the Pr2 simulation perhaps there they have a similiar problem

Hi,

I'm tunning the navigation stack for my robot, but nafvn is displaying a strange behaviour, sometimes it cannot plan when in narrow places.

Here are some screenshoots

In this case the navfn cannot plan

Anybody has an idea why this is happening

Thanks

Hi,

I'm trying to import a wheel collada file made in SketchUp but when I try to atach it to a joint the axis center is in one fo the edges of the wheel. How can I set axis in the center of the wheel so It can rotate. I've already done this with a cylinder instead of a wheel.

Thanks in advance

Hi,

I'm trying to build a system utilizing gmapping and laser_scan_matcher ( laser_scan_matcher to improve the odometry of my robot). I've already made the system work utilizing the bag that came with laser_scan_matcher. Using

Fixed_Frame: odom Base_Frame: base_link

But when I try to use the same setup in conjunction with stage, a concurrent tf transform problem arises.

I kind of understand the problem the laser_scan_matcher publishes the odom to base_footprint and stageros does also.

But I'm lost how to solve it i tried to modify stageros to not publish the transform but laser scan matcher needs the inicial transform.

Thanks in advance

What kind of covariance do you have in both the wheel odometry and imu?

I get that. But I have to build the node that advertises/create the service definition in all of the computers that use that service?

Hi, I've used ROS since the electric version and made several nodes that included services. Now in groovy (catkin) I've tried to make a node that advertises a service, everything worked fine until I called it from another machine and got a unknow service error message.

So I tried to do the tutorial of creating a simple service(AddTwoInts), and the same thing happend.(ok in same machine does not work if called from another) But if I run the cpp begginer tutorials already compiled in ROS groovy the service works in the same machine as well as in multiple machines. I've altered the package.xml and the CMakelist to the same as in the begginer tutorials in ROS repository.

So I'm kind of lost anybody had the same kind of problem?

Thanks, Francisco Marques