Job Description

PhD Title: Advanced Statistical Signal Processing for Next Generation Trajectory Prediction

Summary: Full-time job contract to perform doctoral research in aircraft trajectory prediction. The PhD candidate will join a multidisciplinary team and will be jointly supervised by faculty from UPC and ISAE-SUPAERO. Eventually, the candidate will have the opportunity to collaborate in SESAR (Single European Sky Air Traffic
Management Research) funded projects and/or in cooperation with industrial partners in the field. The main goals of the PhD thesis are: i) The formulation, characterization and assessment of new statistical signal processing techniques for aircraft trajectory prediction, ii) The study of new on-board cooperative trajectory prediction statistical signal processing architectures for future self-separation and conformance monitoring techniques, and iii) The derivation of optimal decision conflict detection and resolution metrics.

Description: Accurate and reliable trajectory prediction (TP) is fundamental for the design of next generation of air traffic services (ATS) decision support tools for traffic synchronization and separation management; as well as enhanced safety nets and collision avoidance tools; either in a (partially) automated environment, on ground, airborne or in a distributed system. Current state-of-the-art technology mostly rely on heuristic decision rules and/or using simplified dynamic models, together with filtering techniques which consider simplistic assumptions and are not designed to cope with system modeling inaccuracies, the latter implying a lack of robustness, known to be of capital importance in safety-critical applications such as for Air Traffic Management (ATM). In this PhD, we propose to look at the TP problem from a new probabilistic perspective, approaching it with powerful mathematical tools arising from the statistical signal processing field. Such advanced robust statistical inference techniques have been shown in other contexts to provide a remarkable performance/robustness improvement with respect to conventional approaches, then probably leading to significant contributions in the TP field. The PhD study will be based on an interdisciplinary approach combining accurate weather and aircraft trajectory modelling and estimation techniques.

Mobility Schedule: This thesis seeks to overcome the aforementioned limitations, and deeply study the design and use of nonlinear Bayesian inference techniques, multi-object/multi-sensor filtering, noise statistics estimation, Bayesian non parametric solutions and Bayesian detection strategies, all of them targeting different challenges within the TP problem, towards a complete probabilistic TP framework. The required interdisciplinary setting will be ensured by the joint supervision from UPC, located in the Castelldefels Campus (close to Barcelona, Spain); and ISAE-SUPAERO, located in Toulouse, France. The candidate will be mainly based at UPC, but a comprehensive mobility plan is proposed with several secondments at ISAE-SUPAERO and at Northeastern University (Boston, MA, USA). Moreover, links with the industry will also be sought, such as Harris, Tern Systems Iceland or ENAIRE. The candidate will also have the opportunity to attend several international conferences and summer schools.

Requirements of the candidate:
• BSc/MSc degree in Telecommunications, Computer Science, Mathematics or related fields.
• Solid mathematical background and outstanding academic records.
• Knowledge on signal processing techniques.
• Strong algorithmic, programming and computational problem-solving skills.
• Excellent communication skills in oral and written English
• Motivation to conduct high-quality research, including publishing the results in relevant venues and conferences.
• Open-mindedness, creative thinking, strong integration skills and team spirit.
• Knowledge on Bayesian filtering (Kalman and particle filters) is a plus.
• Knowledge or personal interest and motivation for avionics/aeronautics is a plus.
• Knowledge in Matlab, Python, R and C/C++ is a plus.
• Knowledge in statistical analysis and machine learning is a plus.

Follow @eurasip