Project LU1: Efficient Learning of Dynamical Systems

Project LU2: Event-Based Information Fusion for the Self-Adaptive Cloud

Project LU3: Secure Software Update Deployment in Autonomous Systems

Project LU4: Design, Optimization and Control of Self-Driving Networked Systems

Project LU5: Testing Autonomous Control-Based Software Systems

Project LU6: Deep People: Learning Integrated Visual Human Sensing Models

Project LU7: AI Reasoning for Situation Understanding in Human-Robot Interaction

Link To Application Portal:  

Application deadline: June 15 2017

Project LU1: Efficient Learning of Dynamical Systems

Supervisor: Professor Bo Bernhardsson, Department of Automatic Control

Contact info: Phone +46 46 2228786, Email: bob@control.lth.se

Short Project Description: The project aims to develop methods that learn complex dynamical models through algorithms that actively explore the behavior of the system. The key challenge is to trade the cost of “exploration” with the future benefits from the resulting improved knowledge. This area, sometimes named “dual control theory”, is now vitalized from recent progress in machine learning and statistical estimation. There is also an increased number of exciting applications where the problem is central, such as in self-driving vehicles and self-learning robots. The applicant should have an interest to work cross-disciplinary in the areas of control theory, signal processing, machine learning and statistics.

Project LU2: Event-Based Information Fusion for the Self-Adaptive Cloud

Supervisor: Associate Professor Anton Cervin, Department of Automatic Control

Contact info: Phone +46 46 2224475, Email: anton.cervin@control.lth.se

Short Project Description: Successful self-adaptive resource provisioning in the cloud relies on accurate tracking of workload variations and timely detection of changes in the infrastructure. The general estimation problem is very challenging due to the massive number of observable events in various subsystems, each containing some useful information. In this project, we will develop novel, event-based estimation techniques for information fusion in cloud server systems. Our starting point will be the family of Monte Carlo-based inference methods known as Particle Filters, which will be adapted to handle event-based measurements from different sources and with different time scales. The results will enable more responsive and exact decision making in the autonomous cloud. Applicants should have a strong interest in control theory and mathematical statistics.

Project LU3: Secure Software Update Deployment in Autonomous Systems

Supervisor: Associate Professor Martin Hell, Department of Electrical and Information Technology

Contact info: Phone +46 46 2224353, Email: martin.hell@eit.lth.se

Short project Description: New software vulnerabilities are disclosed on a daily basis and in order to keep devices and systems secure, deployed software must be updated and patched on a regular basis. For an autonomous system, this poses particular problems since the devices are often unmanned, dynamic, of large-scale, and must adapt to changes. The project will focus on the software update process for autonomous systems. The project will develop and analyze methods and tools for efficient and accurate evaluation of potential vulnerabilities in the context of an autonomous system. It will also analyze protocols and methods for deployment and protection of the deployed software updates. This includes digital signatures and analysis of PKI solutions suitable for a dynamic and adaptive environment. To this extent, blockchains can be used as a distributed ledger for CA certificates. Integrity protection and confidentiality protection as well as DoS mitigation techniques will be considered, making the update process secure, reliable, and robust.

Project LU4: Design, Optimization and Control of Self-Driving Networked Systems

Supervisor: Professor Maria Kihl, Department of Electrical and Information Technology,

Contact info: Phone +46 46 222 9010, Email: maria.kihl@eit.lth.se

Short Project Description: The topic of this project is in the area of design, optimization and control of self-driving and autonomous networked systems. Self-driving networked system are envisioned to form the essential infrastructure for mission-critical services in IoT and edge cloud scenarios. There can be a wide range of application scenarios in the area of mission-critical services, for example smart cities, autonomous vehicles, cloud robotics, eHealth, etc. The project requires excellent programming skills and competence and interest in mathematics, telecommunications, control theory, machine learning, and system modelling.

Project LU5: Testing Autonomous Control-Based Software Systems

Supervisor: Associate Professor Martina Maggio, Department of Automatic Control

Contact info: Phone +46 46 2224785, Email: martina.maggio@control.lth.se

Short Project Description: Self-Adaptive software usually comprises the software itself and an adaptation layer, in charge of observing the current execution conditions and reacting to these conditions with changes in the software behavior. The adaptation layer is often realized with control-theoretical techniques, to exploit the large set of guarantees that control-based adaptation provides. Properly testing these systems is a complex problem. First, the control strategy should be verified on its own to assess the formal guarantees that it entails. Second, it should be possible to verify that the introduction of control theory does not influence the behavior of the software in terms of functional properties. Third, the formal guarantees that the control-theoretical adaptation offers should be verified in practice when the controller is connected to the software system. The project proposes the study of testing for self-adaptive software where the adaptation layer is based on control-theoretical principles.

Project LU6: Deep People: Learning Integrated Visual Human Sensing Models

Supervisor: Professor Cristian Sminchisescu, Department of Mathematics, Faculty of Engineering, Lund University

Contact info: Phone +46 46 222 34 98, Email: cristian.sminchisescu@math.lth.se

Short Project Description: The topic of this project is to explore visual human sensing methods based on large scale weakly supervised deep learning techniques. We seek advanced spatial and temporal models that integrate person localization, pose estimation, as well as action and intent recognition based on images and video data.  Applications are broad, including robotics (human-robot interaction, lifelong learning, protection and security, surveillance), entertainment and virtual reality, content-based indexing of visual libraries and archives, among others.

Project LU7: AI Reasoning for Situation Understanding in Human-Robot Interaction

Supervisor: Associate Professor Elin A. Topp, Department of Computer Science

Contact information: Phone: +46 (0)46 222 4249, Email: Elin_Anna.Topp@cs.lth.se

Short Project Description: When autonomous systems, or robots, interact with humans in an open-ended world (the real world), there will be situations where the system works according to its technical specifications, but its behaviour is incomprehensible for the user. The PhD project will thus investigate how a better situation understanding can be provided for systems that interact with humans in a mixed-initiative setting. Example problems include how such an interactive system can detect ambiguities in the user behaviour, how it can assess unexpected and ambiguous situations, and how suitable responses can be computed. Such responses can mean to act autonomously, or to ask the human for help to resolve an ambiguous situation. Machine learning and pattern classification techniques are assumed suitable for the identification of patterns in user behaviour, while the situation assessment and computation of appropriate actions and responses will be based on reasoning mechanisms, potentially including defeasible reasoning. Testbeds and opportunities for the demonstration of results include different mobile platforms, classic industrial manipulators, and collaborative robots available in the LTH Robot Lab, as well as the WASP demonstrator arena for Public Safety (WARA-PS).