Run on Target Hardware

Detect and pick an object using algorithms deployed on onboard computer using Robotics System Toolbox™ Support Package for Kinova^® Gen3 Manipulators

To complete the overall workflow as part of Model Based Design, the Robotics System Toolbox Support Package for Kinova Gen3 Manipulators enables connectivity and control of the robot using a stand-alone compute board like NVIDIA^® Jetson™. This feature helps you to eliminate the need of a continuous communication link between the robot and a generic host computer (with MATLAB^®). With onboard deployment, you can design various automated tasks involving artificial intelligence, computer vision, and path planning, and deploy the algorithm on compute board to achieve higher efficiency and accuracy. The code generation and deployment requires ROS Toolbox.

Using this capability of the support package, you develop an object detection algorithm to identify object’s position and orientation from an image, localize the object in 3-D space, compute desired trajectory for a manipulator robot to reach the object from current position, and command the manipulator to move accordingly.

Topics

Configure NVIDIA Jetson Board for Object Detection and Motion Planning
Initial configuration to deploy ROS node on NVIDIA Jetson.
Object Detection and Motion Planning Application with Onboard Deployment
Choose the workflow for deploying motion planning applications to onboard computer.
Detect Position and Orientation of a Rectangular Object Using OpenCV
Use OpenCV for proper object detection from vision module of robot.
Joint Angle Control Versus Trajectory Control
Compare the control strategy that you need to adopt for robotic manipulators.

Featured Examples

Simulate a Detect and Pick Algorithm Using OpenCV Interface and Rigid Body Tree Robot Model

Simulate a rigid body tree workflow for KINOVA® Gen3 to detect an object using OpenCV Interface, and then pick and move it with inverse kinematics.

Open Live Script

Design and Validate Object Detection and Motion Planning Algorithms Using Gazebo, OpenCV, and MATLAB

Design and validate complex application workflows for Kinova Gen3 7-DoF Ultralightweight Robot using Gazebo, OpenCV, Robotics Systems Toolbox™ and ROS Toolbox.

Open Live Script

Read Current Joint Angle of KINOVA Gen3 and Publish to ROS Network

Deploy Simulink^® algorithms as standalone ROS node to NVIDIA Jetson compute boards and read current joint angle.

Open Model

Detect and Pick Object Using KINOVA Gen3 Robot Arm with Joint Angle Control and Trajectory Control

Deploy object detection and motion planning workflows developed in Simulink as standalone ROS node to NVIDIA Jetson compute boards.

Open Model