Estimation And Tracking Algorithm For Autonomous Vehicles And Humans

Autonomous driving systems have experienced impressive growth in recent years. The present research community is working on several challenging aspects, such as, tracking, localization, path planning and control. In this thesis, first, we focus on tracking system and present a method to accurately track a moving vehicle. In the vehicle tracking, considering the proximity of surrounding vehicles, it is critical to detect their unusual maneuvers as quickly as possible, especially when autonomous vehicles operate among human-operated traffic. In this work, we present an approach to quickly detect lane-changing maneuvers of the nearby vehicles. The proposed algorithm is based on the optimal likelihood ratio test, known as Page test. Second, we consider another form of tracking: tracking the movements of humans in indoor settings. Indoor localization of staff and patients based on radio frequency identification (RFID) technology has promising potential application in the healthcare sector. The use of an active RFID in real-time indoor positioning system without any sacrifice of localization accuracy is intended to provide security, guidance and support service to patients. In this paper maximum likelihood estimation along with its Cramer-Rao lower bound of the locations of active RFID tags are presented by exploring the received signal strength indicator which is collected at the readers. The performance of real-time localization system is implemented by using an extended Kalman filter (EKF).