WASP at the Department of Computer and Information Science (IDA)

About WASP at IDA

One of the research environments connected to WASP - Wallenberg AI, Autonomous Systems and Software Program at Linköping University (LiU), is located at the Department of Computer and Information Science (IDA) on Campus Valla in Linköping.

At the Department of Computer and Information Science, research is carried out in the fields of artificial intelligence; human-computer interaction; software and computer systems; and data science.

On this page we present more information on the research activities linked to WASP at IDA and the researchers involved. You will find information about:

  • WASP AI and integrated systems
  • WASP Machine learning
  • WASP Real-time systems
  • WASP-ED - a national program for educational development in AI

Publications within WASP at IDA

To publication repository

WASP at LiU

For more information about research within WASP at LiU, go up one level in the menu to WASP - Wallenberg AI, Autonomous Systems and Software Program. At this level, you will also find a contact list of all WASP researchers at LiU divided by department.

 

News

WASP at IDA

A part of the supercomputer that glows red.

AI may boost society's cybersecurity

The race between hackers trying to crack systems vital to the functioning of society and cybersecurity experts is constantly ongoing. LiU-researchers will now develop AI that can detect hacker attacks and take action before any damage is done.

image cut from thesis cover

A new doctor in Computer Science: George Osipov

On the 3rd of June, George Osipov successfully defended his thesis in the intersection of theoretical computer science and artificial intelligence (AI), with a focus on parameterized complexity of constraint satisfaction problems (CSPs).

Image of diploma.

Best paper award in automated planning

David Speck and Daniel Gnad at the Department of Computer and Information Science have won the Best Paper Award at the Conference ICAPS 2024, one of the most important conferences for automated planning, a topic in artificial intelligence.

Piotr Rudol with one of the drones at Gränsö

Work at WASP

Check our our job openings at WASP, Wallenberg AI, Autonomous Systems and Software Program, at Linköping University and apply today.

WASP AI and integrated systems

The main theme of research in the Artificial Intelligence and Integrated Computer Systems Division (AIICS) at IDA, is at the intersection of Artificial Intelligence and Autonomous Systems. The group has over 20 years of experience in this area.

Our research focus is in cognitive robotics. Cognitive robots are systems containing com- putational equipment that provide them with capabilities for receiving and comprehending sensory data, for generating models, for reasoning with such models, and for rational action in their environment.

An equally important area of focus is with collaborative robotic systems. Such systems contain teams of heterogeneous robots and humans that interact and act together in achieving joint goals. Underlying this research is the development of integrated systems of systems and their testing and application in the field. A specific area of expertise is in the development of collaborative unmanned aircraft systems. A specific area of application is with emergency rescue operations.

Artificiell intelligenceArtificiell intelligence Photo credit Göran Billeson

Research projects WASP AI and integrated systems

Simultaneous coalition structure generation and task assignment

In this project, we consider simultaneously generating coalitions of agents and assigning the coalitions to independent tasks. This optimization problem has many real-world applications, including forming goal-oriented teams of agents.

Simultaneous coalition generation and task assignmentSimultaneous coalition generation and task assignment

A fundamental problem in artificial intelligence is how to organize and coordinate agents to improve their performance and skills.

We have developed several state-of-the-art-algorithms to solve the problem. To evaluate the algorithms’ performance, we extend established methods for synthetic problem set generation, and benchmark the algorithm using randomized data sets of varying distribution and complexity. We also apply the algorithm to solve real problems in a major commercial strategy game, and show that the algorithm can be utilized in game-playing to coordinate groups of agents in real-time.

The algorithm solves real problems in a major commercial strategy gameThe algorithm solves real problems in a major commercial strategy game

WARA PS

WASP/WARA Public Safety Demonstrator: WARA PS

Members of AIICS are participants in the WASP/WARA Public Safety Demonstrator providing autonomous systems expertise, software infrastructures and unmanned aerial vehicle (UAV) platforms for the demonstrator.

Artificiell intelligens

WARA demonstrator 2018
FotoPhoto: Thor Balkhed

Equipment and labs

AIICS has a fully equipped Unmanned Aircraft Systems Lab with a Vicon real-time tracking system and a fleet of autonomous vehicles ranging from mid-size Yamaha RMAX helicopter systems and DJI quadcoptors to in-house manufactured micro-UAV systems. The robotics fleet also includes a HUSKY UGV.Artificiell intelligens
Emergency Rescue Mission: Creating adhoc mesh networks for initial communication using collaborative UAVs.

WASP Machine learning

The Division of Statistics and Machine Learning develops the machine learning methods that autonomous systems use when learning from data. Our research and education focuses on probabilistic modeling and efficient algorithms for learning, prediction and decision making under uncertainty.

The researchers at The Division of Statistics and Machine Learning, STIMA, are active in both basic machine learning methodology as well as in applied industrial research in areas such as transport systems, telecommunication, robotics, autonomous vehicles, medical technology, climate modeling, and software development.

The research at the division has resulted in efficient simulation-based inference algorithms for large scale data, new graphical models for describing structure and dependencies in complex data, methods for analyzing causal relationships, and other contributions to the machine learning area.

Machine learning Photo credit iStock/monsitj

Research projects WASP Machine learning

Machine learning for 5G System Control and Automation

TelekommunikationerPhoto iStock/12521104 In comparison to 4G, there is a plethora of new applications envisioned for 5G, ranging from traditional efficient broadband communication links to more strictly reliable communication links suitable for critical communication in industrial applications or massive communication that is typically associated to the Internet of Things.

In this project we develop solutions to increase the autonomy of 5G networks with the goal to improve network performance, reduce operation complexity, and increase resilience.

Employing probabilistic machine learning models in the project is a ground for an optimal decision making, in particular to satisfy the ultra-reliability requirements where uncertainty quantifications are crucial. The project aims to use probabilistic classification and regression frameworks to guarantee flexibility and autonomy of 5G components, and also use spatial and temporal models in order to take into account dynamic aspects such as device velocity, device movement path or daily/monthly variations of the network workload.

To autonomously adapt to various environmental changes and disturbances, the project aims to embed online learning components into our methods. Efficiency and scalability are challenges in the efficient development of the autonomous 5G environment, and development of appropriate big data processing methods is yet another aim of the project.

Publications

External partner

Ericsson AB

Bayesian Learning for Spatiotemporal Processes in Transportation

Transporter Foto iStock/Wenjie Dong The rapid deployment of streaming sensors have made spatiotemporal data increasingly common. In this project we develop probabilistic models for spatiotemporal data with applications in the field of transportation.

The main focus is on developing computationally efficient Bayesian methods for learning, prediction and decision-making for spatiotemporal network-structured data.

Subprojects

  1. A Bayesian Dynamic Stochastic Block Model for Large-Scale Multilayered Networks with Applications to Airline Network Prediction.
  2. Traffic Flow Modeling and Prediction using Gaussian Processes with Road Topology Structure.

Extern partner

Stockholms Lokaltrafik, SL

Publications

Probabilistic models and deep learning – bridging the gap

Probabilistiska modellerPhoto metamorworks

Probabilistic models and deep learning are two very successful branches of machine learning, with complementary properties. In this project, we will develop theory and methods related to the interplay between these technologies, enabling us to take advantage of the strengths of both types of methods.

Probabilistic models—where unobserved variables are viewed as stochastic and dependencies between variables are encoded in joint probability distributions—are widely used in the areas of statistics and machine learning.

Probabilistic models come with many desirable properties: they enable reasoning about the uncertainties inherent to most data; they can be constructed hierarchically to build complex models from simple parts; they provide a natural safeguard against overfitting; and they allow for fully coherent inferences over complex structures from data.

Deep learning is a different branch of machine learning which has recently had remarkable success in a range of different applications related to computer vision, natural language processing and more. In most cases, the deep neural networks (DNNs) are trained discriminatively using supervised learning, with large sets of (annotated) training data.

Unfortunately, the advancement of deep learning has come at a price—DNNs often lack many of the desirable properties of probabilistic models, such as uncertainty quantification and structure exploitation over well-defined probabilistic priors.

In this project, we will develop theory and methods related to the interplay between probabilistic models and deep learning. We intend to develop both new models and new inference and learning algorithms for applications where unobserved variables are naturally characterized using, for instance, probabilistic graphical models or stochastic processes, whereas data is from some domain where deep learning has been successful (e.g., images).

We will consider applications to clinical decision support, active learning, weak supervision, and semi-supervised learning in the context of dynamical systems. However, the family of problems that involve such interplay between deep learning and probabilistic models is general, and we expect the results from the project to be widely applicable.

External partner

Chalmers University of Technology:

  • Lennart Svensson (PI)
  • Jakob Lindqvist (WASP-MLX PhD student)

WASP Real-time systems

Research at Real-Time Systems Laboratory covers the dependability and timeliness of distributed systems. We conduct research into dependability attributes such as safety, reliability, availability, and integrity. The latter two are also hallmarks of security.

Dependable systems are those for which reliance on the system’s services is justified.

Our research covers the design of distributed algorithms and protocols in networked systems, and methods for ensuring system dependability in the presence of faults, overloads and attacks. The lab has more than 20 years of experience in several application areas such as avionic systems and intelligent transportation systems (ITS). We have also participated in European research cooperation in the area of critical infrastructure security since 2001.

Current projects and activities concern, for example, security in vehicular networks, resource management in edge computing, conceptual design for safety and security in avionics, time determinism and multicore computations, ensuring safety in systems that include learning-based components (ITS and aerospace), risk analysis, and intrusion detection in industrial control systems.

The latter is performed in the context of a 5 year national program RICS (www.rics.se), financed by the Swedish Civil Contingencies Agency (MSB), in cooperation with the Swedish Defence Research Agency (FOI), to create a national testbed for security in Supervisory Control and Data Acquisition (SCADA) systems.

Real time securityReal time security Photo credit oonal

Research projects WASP Real-time systems

Verification of Safety-critical and Learning-based Software

Datasäkerhet
The project aims to extend today’s methods for assurance of safety-critical system to apply in future systems with machine learning components.

Autonomous systems with machine learning components will inevitably be used in environments that can potentially harm humans or the environment.

The project will study formal verification techniques and develop novel methods that provide evidence that such future systems behave as intended. Among properties of interest are robustness, decisiveness and correctness with respect to the intended function.

External partner

Saab group

Explaining outcomes of deep learning systems

DjupinlärningSymbolic representation In this project we aim to understand and analyse deep learning systems and provide explainable results. Specifically, we explore a new technique of symbolic representation of abstractions.

Deep learning systems have three characteristics that makes them difficult to trust in critical applications. Being statistical, they can not be deployed in contexts when worst case performance needs to be relied on. Their results can be difficult to interpret and come with no explanations, and they are notoriously fragile.

In this project we explore a new technique of symbolic representation of abstractions. We will build new tools to verify effectiveness of the method on an autonomous vehicles vision perception system.

External partners

Professor Carl Seger, Chalmers Institute of technology and professor Liu Yang, NanYang Technical University Singapore.

WASP-ED

In January 2022, WASP-ED was launched, a national program for educational development in AI. The goal is to significantly increase Swedish universities’ capability and capacity to provide current, relevant education in AI. The host university for the initiative is Linköping University (LiU).

WASP-ED was launched by WASP - Wallenberg AI, Autonomous Systems and Software Program together with WASP-HS (HS stands for humanities and society). Through WASP-ED, all the higher education institutions that are connected to WASP or WASP-HS will be able to jointly gather strength to spread knowledge about AI so that more education programs adapt elements that include AI. It will be both a development and research program that complements the research done within WASP and WASP-HS.

There are many technical education programs that include AI courses, but the purpose is to also introduce AI in other programs. For example for those who will become HR specialists, economists, doctors, nurses, teachers and more to understand how AI affects their profession and how they can benefit from AI.

National website for WASP-ED: https://wasp-ed.org/

All researchers within WASP at IDA