Reactive range-free landmark navigation without scene reconstruction
Bekris, Konstantinos E.
Kavraki, Lydia E.
Master of Science
Popular approaches for mobile robot navigation involve range information and metric maps. For many sensors, however, such as cameras, the angle between two landmarks is easier to measure. With these sensors' features in mind, this thesis presents an angle-based control law, which allows a robot to reach a subset of the plane by monitoring the bearings of three landmarks. By analyzing this law's properties, a second law has been developed that reaches the complementary set of points. The two methods are then combined in a novel sensor-based path-planning framework that reaches any possible goal in a planar obstacle-free workspace. Experiments have been conducted both on a real robotic platform equipped with a panoramic camera that show the efficiency and accuracy of the proposed techniques. This work provides evidence that navigational tasks can be performed using only a small number of primitive sensor cues and without the range information or scene reconstruction.