Novel exproprioceptive sensor configurations and user interfaces for mobile robot control in uncertain environments

Modern mobile robots are equipped with numerous sensors for proprioception (monitoring internal state) and exteroception (detecting environment state). However, little work has been done on exproprioceptive sensors, which that detect the pose of a mobile robot relative to its environment. Robot operators need a clear visualization of the scene to navigate a mobile robot efficiently. However, sensors that facilitate exproprioception have conventionally been either too expensive or less efficient. In this project, we have explored the use of emerging low-cost depth sensor, Microsoft Kinect to enhance the remote situational awareness of robot operator. A full-featured prototype, Kinect Sensor Enabled Navigation System (KSENS) was built to demonstrate the key principle behind the sensor configuration. Experiments were conducted to verify the efficiency/limitation of KSENS under different scenario and the feedbacks were obtained for future enhancement. Our experiment results illustrated that KSENS is particularly useful when the lighting condition is poor or image resolution is impoverished. Kinect sensor has proven to be beneficial in navigation. However, limited field of view and close range blind spot of Kinect sensor should be resolved in order to maximize its usefulness in the field of robotic.Bachelor of Engineering (Computer Engineering