Ashesh's profile - activity

Actually, I guess this doesn't help. As soon as one starts interacting with the robot via Rviz, the controllers are stopped. One can run rosrun pr2_controller_manager pr2_controller_manager, to get the status of controller.

Okay, I figured out the answer to this myself. What happens is, ROS interactive manipulation stops the controllers. So in order to regain control of PR2 one should restart the controllers using the following command:

rosrun pr2_controller_manager pr2_controller_manager start <controller_name>

In my case the controller_name = r_arm_controller

Hello Everyone

Is there any way to regain control of PR2 after interactive manipulation, in simulation, using Rviz?

The problem I am stuck with is, I interact with robot's arms using "interact" tab in Rviz interactive manipulation and then try tucking them using pr2_tuckarm.py, but the robot doesn't respond to it. Even after closing Rviz interactive manipulation (both robot.launch and desktop.launch) I am not able to tuck robot's arms. The only solution I currently have with me is relaunching pr2_gazebo, but I am sure there must be a better way of doing it.

Hey

Is it possible in ROS to plan simultaneously for right_arm and torso lift joint? During plan we need to set the group name in the following manner:

goalB.motion_plan_request.group_name = "right_arm"

But this plans only for right arm i.e. 7 dof. What should be the group name if I want to plan together for right arm and torso lift joint?

Is it possible to do path planning in ROS (only during simulation) without using any sensor data (eg. kinect, laser or stereo)?

Since the environment data is stored in collada format through which we have a perfect knowledge of the environment (even location which are not visible to PR2) and this data can be used for checking collision. I think this is how OpenRave also checks for collision while planning and it bypasses the sensor.

Hello

I am trying to generate diverse set of trajectories for doing a given task (e.g. moving an object from A to B). Currently I am generating such a set of trajectories by changing goal bias of RRT algorithm. All the trajectories obtained using this method have different number of points and hence they should be different. But after passing them through trajectory_filter all the trajectories becomes same and with same number of points.

From what I understand trajectory filter (CubicSplineShortCutterFilterJointTrajectoryWithConstraints) tries to both smoothen and shorten the trajectories. Is it possible to just obtain a smoothen trajectory without any shortening? I can't by-pass trajectory_filter because I also need the time and velocity information.

Moreover, I am not able to get the difference between trajectory_filter and joint_trajectory_generator. It appears to me that both servers almost the same purpose. But I guess joint_trajectory_generator onny smoothens the trajectory and doesn't do any kind of trajectory shortening.

Any help would be appreciated!!!

Thanks!!!

Hello

How can I save PR2 arm trajectory and then replay the same trajectory at a later time? One way I can think of it is by writing trajectory_msg::JointTrajectory to a file and then reading the trajectory from the file whenever I need to reproduce the trajectory. But if I follow this method then I am not sure how to save "duration time_to_start" (which have many attributes) and "Header header".

Does anyone knows of any neater way for saving/replaying trajectories?

Thanks!!!

Is it possible in ROS to save a trajectory and then play it at a later time? One way I can think of is writing trajectory_msgs/JointTrajectory to a file and then reading it when I need to execute the trajectory. But I am not sure how to save duration time_to_start. Duration seems to have too many attributes.

Hello

I am trying to modify the implementation of RRT in ROS to adapt it for my problem. I am currently using Fuerte. But I am not able to find RRT.cpp anywhere in the ROS insallation in `/opt/ros/fuerte/` (but I could find RRT.h). I also tried the following:

    find -iname '*RRT.cpp*'

but got no search result. Then I started looking into the file ompl_ros_planning_group.cpp
and found that the planner is being called on line #446:

      bool solved = planner_->solve(request.motion_plan_request.allowed_planning_time.toSec());

I am unable to find this solve() function anywhere in the ROS installation. Can someone please help me with this.

I have also seen the ompl code available at ompl site. Where the solve() function was present in RRT.cpp file. 

Thanks in advance!!!

I am facing a similar problem....