Requirements and Challenge

Requirements for a Passing Project Submission

In the Pick and Place Simulator - Test mode, a target object may spawn in any of the nine possible locations on the shelf. For a passing submission, your code must succeed 8/10 times performing a complete pick and place operation. Note: Please include a screenshot of the completed pick and place process!

A run will be considered valid if your code results in a trajectory that is slightly off but if the trajectory is wildly off, that run will be marked as invalid or failed even if pick and place operation was successful.

Since the rviz window displays the actual path planned by the simulator, you can easily verify if your code has generated an acceptable result or not.

As the project is setup with automatic reaching and grasping motions, your submission will fail only if your IK calculations were wildly off. Hence, once you have performed the kinematic analysis correctly, it should be relatively easy to arrive at a passing submission.

If you're struggling to with parts of this project, or just want some help getting your code up to the minimum standards for a passing submission, we've recorded a walkthrough of the basic implementation for you but spoiler alert: this Project Walkthrough Video contains a basic solution to the project!. Please give the project a try before watching this video!

Optimum Performance Challenge:

To obtain the bragging rights of a standout submission, you must calculate and plot the error in end-effector position generated by your joint angle commands.

You can do this by calculating the end-effector position via Forward Kinematics using your code output (which is a set of joint angles) and comparing it with the end-effector position you received in the initial CalculateIK service request.

Have a feeling your error curves are the best? Check out how others did in the #pick_place slack channel and share your results to find out who is the IK champion!