iRoboBumper: SONAR-Obstacle Detection System for Robotics Platforms
Licensed under CC BY-SA 4.0
Summary
This project for the Invisible Robotics Bumper (iRoboBumper) aims to create a SONAR-based obstacle detection system for robotics platforms that provides a full 360 degree horizontal coverage while maximizing the update rate. The self-contained unit includes a base with screw holes for mounting onto a robot chassis, is powered by four AA batteries, and communicates with the host robotics system wirelessly.
Background
As humans we do not generally give much thought to how our brains process the data coming from our senses to provide us with information. When we walk towards a door to leave our house, we do not consciously think, “I am 10 feet from the door; now I am 6 feet from the door; now I am 2 feet from the door; stop!” However, our brains are using data from our eyes to approximate distance, as well as a learned understanding of how far we need to be from the door in order to reach the doorknob. Even this statement over simplifies the processes that are at work though, as the approximation of distance to the door includes such processes as depth perception from the parallax between our two eyes and many monocular cues that aid in estimating distance. Thus, emulating this behavior in the actual way that humans sense their surroundings is an extremely challenging problem.
If however, you want to emulate human behavior in a robotics platform, you need to employ sensors capable of providing the raw data that can be used to calculate the same type of information. Our eyes operate in the visual light spectrum (i.e. ~ 400-700nm) and our brains essentially use complex algorithms to estimate distance. Luckily, there are simpler ways to determine distance using electronics than using a camera and image processing to determine the distance to objects. One technique is with Sound Navigation And Ranging (i.e. SONAR), a system that emits pulses of sound and listens for echoes as they reflect off of objects. The distance to the object can be determined by using the time it takes for these pulses to reach an object and reflect back.
We would love to hear what you think about this project; please tell us in the comments section below.
