Carnegie Mellon University
Abstract:
With continuing advancement in technology, the future of planetary exploration is likely to be dominated by robotic missions. Yet rovers capable of science investigations are slow and bulky with very limited computing which prohibits demonstrating full autonomy. These rovers are also risk averse due to their huge mission cost. However there is a new breed of micro-rovers which are nimble and carry enough computing onboard to perform complex tasks like autonomous navigation. These rovers are also intended to be more expendable as multiple copies of these can be deployed in the same mission and hence can afford to take more risks.
Terrain perception is an essential component of autonomous navigation. Stereo cameras provide a superior solution due to their weight and reliability. However they are limited by the available light. When operating in darker environments like lunar poles, structured lighting provides an efficient solution by focusing the light where it is needed. The structure of the light can also be exploited to understand the geometry of the terrain.
In this talk we will look at how structured light can be used to map a challenging lighting environment. Specifically we will look at different methods of using laser line stripers in the context of terrain mapping. We will also consider the inherent challenges specific to micro-rovers. We will then propose a new technique which uses passive actuation to attain the same mapping capability as actuated methods. We will also explore using the structure of the light and by exploiting the stereo setting, formulate a method to refine the accuracy of the obtained depth image which can then be used to build the terrain maps.
Committee:
David Wettergreen (Advisor)
George Kantor
Michael Kaess
Paloma Sodhi