Autonomous navigation and terrain sensing
A rover-navigation project combined stereo vision, laser-range and infrared sensors with three-dimensional terrain modeling. The method was developed in MATLAB and tested on a Pekee II robot using the programming interfaces from its manufacturer, Wany Robotics, so that the robot’s movements adjusted to the objects its sensors and camera detected.
The project shows how terrain modeling and sensor-guided movement can work together in autonomous navigation.
Mapping and indoor localization
An indoor mobile-robot project implemented and optimized an RGB-D simultaneous-localization-and-mapping (SLAM) method, in which a robot builds a map of an unknown space while estimating its own position in it. A Kinect camera, which supplies color video plus a depth value for every pixel (the D in RGB-D), provided real-time input for mapping and localization, connecting computer vision with spatial estimation on a mobile platform.
The navigation and SLAM projects address two distinct robotics problems: moving through terrain with sensor-guided decisions, and mapping a space while estimating a robot’s location.
Cooperative robots and embedded communication
A University of North Texas research-experience project involved two autonomous LEGO NXT robots coordinating over Bluetooth. The lead robot followed a path using a light sensor and sent instructions to a second robot, and the report documents how Bluetooth communication and the NXT-G programming software were used to coordinate the pair. Cooperation between small embedded robots is a classic embedded-systems problem: each unit has modest computing power, so sensing, messaging, and decision-making must be shared carefully between them.
A related poster shows how the work was presented, while the Research Experience for Teachers project describes its educational setting.