Using camera for simultaneous localisation and mapping with mobile robot

Hkratna lokalizacija in kartiranje je računalniški problem, pri katerem je cilj popis neznane okolice v karto in istočasno slediti lokaciji robota. Razvoj na tem področju je v zadnjih letih v razmahu in primere aplikacij, ki so povezane s tem vidimo zmeraj pogosteje v vsakdanu, kot so avtonomni avtomobili, droni, sesalci, itd. Zaradi preprostosti sistema je uporaba čim manj senzorjev dobrodošla. S tem namenom in njihove dostopnosti se zmeraj pogosteje uporabljajo kamere. Tako bomo v tem delu pregledali teorijo na kateri slonijo algoritmi za reševanje tega problema, uporaba kamer in robota za raziskovanje okolice. Na podlagi praktične aplikacije teh sistemov smo ugotovili, da lahko z zelo preprostim sistemom uspešno popišemo okolico in sledimo lokaciji robota.Simultaneous localisation and mapping is a computational problem at which the goal is mapping an unknown environment and simultaneously track the robots location. Research in this field is in full swing and examples of applications that involve this problem can be seen ever more frequently in our everyday life as in autonomous cars, drones, vacuum cleaners, etc. For preserving the simplicity of a system the goal is to reduce the number of sensors. With this intention cameras alone are used increasingly more because of their accessibility. In this work we will be discussing the theory on which algorithms for solving this problem are based on, use of cameras and robots for exploration. On our application of these systems we have observed that with a relatively simple system we can successfully create a map of the environment and track the robot location