PRISM Updates: Some of the sensors being pondered by Robotics Team

Speaker: Richard Stansbury, student, EECS, University of Kansas, 2002.

Modified Part One Transcript: We haven't yet figured out completely all of the sensors that we are going to use. We do have a good idea of some of the primary sensors, but many of them will end up needing some sort of back-up system because of the risk of failure.

For positioning we are definitely going to be using differential GPS technology because it is the most accurate location sensor available, and with some of the newer GPS technologies we can get very reliable accuracy in real time, so we will be able to have a very good idea of where the robot is. We will know enough so that we can easily move it to desired locations.

For hazard detection and avoidance we are probably going to be going to use a laser range finder. The laser range finders put out by the SICK corporation are fairly standard for field robotics and they were used successfully for Nomad, so they seem to have a lot of potential.

For velocity and acceleration measurements we are looking again at using more differential GPS. This is because as our position moves, we can actually track the velocity and how well we accelerate. Also, unlike a lot of other sensors such as shaft encoders, this remains reliable despite the harsh environment.

Modified Part Two Transcript: For environmental sensing we are probably going to have some sort of weather station, similar to some of the smaller units that people put on their house or their RV. These instruments can correlate barometric pressure, temperature, wind speed, and humidity. We wouldn't need to sense such things as precipitation, as there is no natural precipitation in Antarctica. We would need something that is all inclusive and that provides a lot of information from one unit.

For general orientation we could use what are called tilt sensors, or inclinometers. These can help us know the pitch and roll of the robot as it travels over the snow and ice. We could use differential GPS again, because using more than one receiver would allow us to extrapolate different kinds of orientation data.

For internal sensing we would probably be going with some digital fuel gauges, and some internal temperature sensors.

For outreach we are looking at things such as pan and tilt cameras that we can use for obtaining pictures, the possibility of a panoramic camera, or a thermal camera. These are all examples of what we might use, but we are still looking at the options and what can actually handle the temperatures for that purpose.