Robotics Challenge

Description

Robotics assignment, western university of applied science.

Next scheduled meeting: Onsdag 23/11-22 Ferdigstille rapport

Previous meetings

• Torsdag 17/11-22 11:15 Final presentation, Zoom
• Torsdag 03/11-22 11:15 CDR, Zoom
• Tirsdag 18/10-22 14:00 Discord
• Tirsdag 11/10-22 14:00 Discord
• Torsdag 6/10-22 11:00 Zoom PDR breif
• Tirsdag 4/10-22 14:00 Discord
• Tirsdag 27/9-22 12:00 Discord
• Tirsdag 20/9-22 14:00 oppstart

Overarching project tasks

• Publish tasks.
• Interface topology (All)
• Kinematic modeling (Mathias)
  • Kinematic model
  • Model Simulation Matlab
• Motion algorithm (Mathias)
• Sensor package (Fredrik F.)
• Make Model(Fredrik F.)
  • Make 3D model of assignment
  • Model simulation Gazebo
  • Rviz
• Robot arm (Fredrik, Lars, Sander)
  • Moveit implementation
• Preplaned Navigation Planning (...)

Project Tasks

Develop the forward kinematics of your robotic solution, in Matlab (not Toolbox) or by hand:

For the robot arm(s):

• Develop the table of DH parameters
• Develop the transformation mapping end-effector to base (for the first 4 joints only)

For the mobile robot platform(s):

• Draw a model of the mobile robot with the necessary variables defined (see Fig. 4.1 in Corke for inspiration)
• Develop the kinematic equations of motion for the mobile robot
• Discuss whether the mobile robot is holonomic or non-holonomic

For the robotic system in general:

• Develop the transformation from the chosen sensor system to the relevant coordinate system on the robot (world, end-effector, mobile robot, etc)

Model your robot kinematics with Peter Corke's Robotics Toolbox in Matlab:

For the robot arm(s):

• Demonstrate equivalence of the forward kinematic solution obtained previously in Matlab (not Toolbox) or by hand
• Develop the differential kinematics (i.e. relating joint and Cartesian velocities), and demonstrate how it could be used
• Develop the inverse kinematics, and demonstrate how it could be used
• Demonstrate example motion planning, on a task relevant to your robot design challenge (or similar)

For the mobile robot platform(s):

• Determine suitable controller(s) to control the mobile robot for your chosen challenge
• Implement the kinematic model and the controller(s) in Matlab(/Simulink)

For the robotic system in general:

• Demonstrate using the sensory system to command the robot, according to the task chosen. That is, show the calculations necessary to make the sensory data (e.g. an apple detected at an arbitrary location from a static 3D camera) useful to the robot (e.g. calculate the joint angles to put the tool point of the end-effector at the apple’s location).

Simulate the kinematics of your robot in Matlab:

For the robot arm(s), depending on robot design challenge either:

• Use motion planning to move the robot end-effector through the required positions/orientations for the task chosen, or
• Use differential kinematics to move the end-effector using velocity commands according to the task chosen

For the mobile robot platform(s):

• Simulate your chosen challenge, and discuss the simulation results in terms of chosen control strategy and performance
• Discuss and implement a navigation strategy for the mobile robot for your challenge
• Discuss how you would implement a localization strategy for the mobile robot for your challenge

Connect the Matlab code to ROS and simulate the physical robot in Gazebo

For the complete system:

• Model your complete robot system using URDF and visualize the robot in Gazebo
• Your robot arm(s) mounted on your mobile platform
• Your mobile platform, with wheels, sensors etc

For the robot arm(s):

• Demonstrate controlling your robot arm(s) in Gazebo over ROS from Matlab, by following along a trajectory calculated in Matlab, or controlled using your differential kinematics implemented in Matlab.

For the mobile robot(s):

• Demonstrate controlling your mobile robot platform in Gazebo over ROS from Matlab.

Optional

• Control a physical UR, Turtlebot, or other robot using coordinates calculated with your Matlab code through ROS.