We invite applications for a fully funded doctoral researcher position in the field of deep learning for end-to-end motion planning of unmanned aerial vehicles.
The project is supported by the H2020 ICT – RIA program OpenDR for research and development in Deep Learning for Robotics. In this project, we will introduce end-to-end motion planning methods for UAV navigation. Informed by a rough path to goal in partially known environments, the developed method will create desirable, local motion plans using raw images from the front-facing camera on quadrotor. According to our scenario, environment is partially known without exact obstacle location information, an initial rough path to goal is given, and concatenation of desirable local motion plans for safe navigation is to be found. Such scenarios can be seen in many indoor navigation problems, such as autonomous drone racing. Contacts: Applicants seeking further information are invited to contact:Assoc. Prof. Erdal Kayacan (er...@eng.au.dk) How to apply: Please follow the instructions here: http://phd.scitech.au.dk/for-applicants/apply-here/november-2019/deep-learning-for-end-to-end-motion-planning-of-unmanned-aerial-vehicles/ In the dropdown menu you must choose the project: Deep learning for end-to-end motion planning of unmanned aerial vehicles (Dlempu) After submission of the application, you will receive a confirmation e-mail with an application ID, you should use for reference if needed.
_______________________________________________ uai mailing list uai@ENGR.ORST.EDU https://secure.engr.oregonstate.edu/mailman/listinfo/uai
