Projects

If you have implemented something using oblu which you could be part of this section, do share with us. Best way to share videos and documents are Youtube and Slideshare respectively. Just send us the links to share and a brief summary of less than 150 words.

Indoor Positioning System For Fire-fighters

  

Take a look at the fire-fighters who take on the blaze from the inside, and the technology that's created to help them. This video covers testing of an indoor positioning system for fire fighters developed by KTH Royal Institute of Technology, using shoe-mounted sensors to aid navigation.

This cooperative localization system comprises of shoe-sensors for localization, UWB radio for short range communication and simultaneous inter-agent rangint and application software for processing, monitoring & sharing realtime location.

The technology will save more lives from fires while making the work of a firefighter safer. A shoe sensor - small embedded computer - in the heal of the firefighter's shoe allows for the leader of the rescue operation to follow the firefighters movements and location throughout the coordinated rescue operation -- without any pre-installed infrastructure like GPS, WiFi, maps etc.

An Amazing Robot With A Cool Smartphone App

Can’t view? Download (6.5 MB)

User draws path on the smartphone's touch display and presses the button to send it to the robot. Robot follows the path and simultaneoulsy transmits it realtime location to the phone. The Robot is based on Arduino and the motion sensor module "oblu". Bluetooth is used for communication between phone and the robot. The robot is making use of oblu's inbuilt bluetooth. There is no separate bluetooth module on robot. Interestingly, it is the pedestrian navigation algorithm on oblu which is helping robot in tracing the user defined path.

Navigation with "oblu"

Can’t view? Download (10 MB)

This video covers demonstration of using Pedestrian Dead Reckoning (PDR) sensors with Google map. Initialization (setting initial position and initial heading) is done manually, using phone's GPS and compass. Phone and the device must be perfectly aligned during initialization. GPS is used for initialization only. Rest of the positioning is performed using shoe (inertial) sensor only.

*Most updated satellite images from other sources confirm that Google map is not very accurate.

Indor Navigation & Localization of a Robot

Can’t view? Download (22 MB)

The robot moves in a pre-programmed path and transmits its current coordinates to a phone for real-time tracking. An Android application Xoblu is used for real-time tracking on phone. The robot's movement is NOT remotely controlled, except start / stop commands. oblu (IMU) senses motion and communicates movement information to Arduino. Based on the programmed path and the motion information (sent by oblu), Arduino controls wheels' movements.

Indoor Pedestrian Navigation With Two Shoe-Sensors

  

John-Olof Nilsson (Researcher, Signal Processing Department, KTH, Sweden) did some ten (10) minutes walks, with each walk covering approximately 700 meters distance around his lab. Same path was traced ten times. The film runs at x20. Two shoe-sensors were mounted on each food and an Android phone was used for collecting and realtime displaying. As always, the positioning was performed using accelerometers and gyroscopes only. Fusion of the two MIMU22BTP sensors took place on the phone. The map was overlay and manually adjusted to the first lap of the trajectory. Tracking was performed using foot sensors only. No other infrastructure (GPS, WiFi, maps) were used.

Robot Navigating a Predefined Path

Can’t view? Download (41.91 MB)

Somewhat excited to share this new experiment in autonomous robotics, performed with the opensource oblu (IMU - inertial measurement unit). Once again oblu controls movements of Arduino based robot. This time we implemented little more sophisticated error correction mechanism. Robot traverses a predefined path, which consists of multiple straight line segments of 0.5 meters. Black crosses on the floor are for marking purpose only. After making ten 90 degree (5 left, 5 right) and one 180 degree turn, robot ends its journey at the start point. Stay tuned for more excitement. Guaranteed.

Pedestrian Navigation Using "oblu"

Can’t view? Download (3.9 MB)

The sensor module, when attached to the foot, tracks the person. No GPS or maps or any other wireless positioning technology like WiFi, Bluetooth etc are used for this demonstration.

Dots correspond to the steps location.

PS: Phone displays GPS coordinates on screen. This information is not used in anyway for this tracking experiment.

© 2016-17 GT Silicon Pvt Ltd. All rights reserved.