Embodied Autonomous Systems

Description

Learning outcomes MMK1 Scientific-disciplinary knowledge and comprehension in the field of Artificial Intelligence for Robotics MMK2 Gaining in-depth knowledge and comprehension, including the ability to develop relevant prototypes or proof-of-concepts, in at least one of the following disciplines in Artificial Intelligence: machine learning, deep learning, knowledge representation, computer vision, audio processing, natural language processing, search and optimisation. Gaining in-depth knowledge and comprehension of at least one of the following application domains of Artificial Intelligence: health, education, logistics, manufacturing, robotics. Objectives After successful completion of this course, a student will - possess knowledge of the key components and terminology adopted in embodied autonomous systems (including a.o. autonomous vehicles, aerial robots and autonomous ships); - have gained insight in the similarities and differences between mobile robots and robot manipulators, and in the kinematic modelling of mobile robots; - have knowledge about sensors typically adopted in mobile robotics such as laserscanners, odometry sensors, and inertial measurement units; - understand the background of and be able to apply basic probabilistic (Bayesian) estimation techniques such as histogram filters, Kalman filters, or particle filters, in particular for localisation and map building (SLAM) purposes; - be able to select and apply an appropriate navigation algorithm for an autonomous vehicle; - be able to solve on paper simple mobile robot localisation and decision making problems using techniques discussed in the course; - be able to use the provided software tools to develop solutions on a computer for more complex mobile robotics applications.