MathWorks announces the winners of the 2016 MATLAB and Simulink Hardware Challenge. Congratulations and thanks to all those who entered.
This project features an omnidirectional wheeled robot that can also move sideways as opposed to normal cars and robots. The complex simulation for this robot is validated by deploying the code to an actual robot. In order to properly control the robot in a dynamic environment, Robotics System Toolbox is used to create a controller that is robust and adaptive. Image Processing Toolbox is also used to determine the real-time position of the robot. By using SimMechanics, the experimental trajectory of the robot is compared with its simulated trajectory. This project illustrates how various toolboxes can be used to methodically modify a simulation to be used in the real world.
This project aims to enhance the virtual reality experience by sending data from various sensors, such as the Microsoft Kinect, to MATLAB so that the information can be processed and sent back to a virtual reality headset. Image Acquisition Toolbox is used to obtain information from the Kinect sensor and then objects of interest are detected by MATLAB. Image Processing Toolbox and Computer Vision System Toolbox are utilized to further analyze these images coming in from the Kinect sensor. Additionally, this group uses serial communication to send data between the various sensors, Arduino boards, and MATLAB. This project demonstrates how several devices and MATLAB can be used to enrich immersive technologies.
This project showcases a hexapod robot capable of making decisions based upon environmental triggers such as colored tags. By using Image Processing Toolbox and Computer Vision System Toolbox, the hexapod reacts to visual stimuli captured by the robot’s on-board camera. The movement of the robot is determined by a control algorithm which utilizes the mathematical model of a neural circuit. To mimic stereovision with only one camera, Curve Fitting Toolbox is used to analyze experimental data. This project is an example of how low cost hardware can be used for machine learning.