Skip to main content

Featured Post

Product Design of a 4 wheel differential drive robot

A Multilayer Neural Network Based Obstacle Avoiding Robot | ROS | Simulation

This is an obstacle avoiding robot which I and Sooraj Krishna built in ROS using STDR simulator. The robot is making decision based on a 3 layer Neural Network. The learning algorithm used is back-propagation (which we wrote in python). The network design is shown below.

The 4 inputs are distances from sonar sensors placed on the Robot. The two outputs are linear velocity on x -axis and angular velocity on z-axis. The Neural Network is first trained using the data captured from joystick controlling the simulated Robot. Then the  trained weights are used in the NN for controlling the Robot. This is the first version that is trained with small training set. The video is showing the Robot avoiding obstacles on its own.

This is after a bit more training.

NB: The video is playing a bit more faster than the actual simulation as you can see from the time showing in simulator.

NB: Among two outputs one is for angular velocity. For turning right the value will be positive and for turning left it will be negative. The network was unable to converge since the output has negative values. I had to shift the range so that all the output values are positive. Need to play with this some more time to figure out the rest. :-)

Comments

Popular posts from this blog

Plymouth theme for Ubuntu

Bored of having the same boot animation screen again and again? There are plenty of Plymouth themes available out there. This is a theme I created by slicing some cool gif files I found online. You can download the theme and find the installation steps on my github account. ( https://github.com/krishnan793/PlymouthTheme-Cat )  The theme is created for Ubuntu 16.04. But this can be installed on previous versions with slight modification. If you have a slow computer then you can watch the whole animation loops. (The VM I used to record the screen was fast though. :))

Getting started with GPIO (Testing GPIO with a multimeter)

There are 65 possible GPIO pins that you can control. Some of the GPIO pins are normally configured to different functions. Let's play with one of them. Go to the beaglebone website and figure out the header pin corresponding to gpio23. The gpio23 is mapped to the 13th header pin. In this tutorial we are testing the gpio23 output using a multimeter (Do not connect an LED directly to BBB IO pin). First go to following directory cd /sys/class/gpio / Then make gpio23 available echo 23 > export A new directory will be created. Open gpio directory. cd gpio23 The direction of IO will be normally set as in (input).  To make it as output, echo out > direction Then set the pin to high using echo 1 > value Measure the voltage at pin 13 of P8 header. If its 3.3 V then you are on the right track. Pin 1 and 2 will be ground. You can connect the negative terminal of multimeter to pin 1 or 2. To set output as zero, echo 0 > value After finished wit...

Remote access your BeagleBone Black using vnc

Before going into how to share desktop with your BBB I assume you have a proper ssh access to your BBB. (through USB) (if not read this ) Also this tutorial is for BBB with Debian installed (type cat /etc/*-release to know which distro you are currently using. Or refer here ) First install the vnc server in BBB.(After logging into BBB) ssh root@192.168.7.2 apt-get install x11vnc Then go to your local computer and open a terminal window. Install a vnc viewer like vinagre. sudo apt-get install vinagre Now its time to start vnc server in BBB. x11vnc -auth /var/run/lightdm/root/:0 -forever This will start a vnc server at port 5900. Note this server setup is temporary. If you want to remotely connect to your BBB after a reboot type the above command again. Go to your local computer and type  vinagre 192.168.7.2::5900 A remote desktop will be shown if all goes well. Try exploring other options in x11vnc using (in BBB) man x11vnc Th...