Utilizing Industry-Academia Collaboration for Better Software
Interviews and coverage
Reportage: Samverkan skapar bättre mjukvara
Eric Giertz, KTH, om samspelet mellan akademi och industri
Lufoss 2016 #theta360 - Spherical Image - RICOH THETALUFOSS-gänget i form av vinnare, arrangörer och jury, anno 2016
The objective of the annual LUCAS-dagen is to inspire and develop collaboration between industry and academia in the field of applied software. The focus of LUCAS-dagen 2016 is to explore the diversified experiences from the last 15 years of mutually beneficial Industry-Academia collaboration. In order to further improve our relations. We have invited Professor Eric Giertz, KTH, to share research results and experiences on the topic. On the technical side Mark van den Brand, Eindhoven University of Technology, will focus on the industry relevant topic of Model Driven Engineering in High Tech Industry.
When we, as usual, present fresh results from joint research projects, we will also reflect on the different forms of collaboration behind the results. On an organizational level, long-term partners such as Ericsson and ABB will share their experiences and expectations – and so will newer members in the innovation system, such as Bosch.
09.30 Registration, coffee, posters & demos
Demos and posters:
- IoT Navigator (Sten Minör)
- Multilingual knowledge transfer using entities (Peter Exner)
- SCALARE - A framework helping industry to scale up software (Martin Höst)
- Demo: DrAST - Compiler Visualization Tool (Jesper Öqvist, Görel Hedin)
- Demo & poster: Langforia: Language Pipelines for Annotating Large Collections of Documents (Marcus Klang)
- OSA - An Open API for Smart Apartment buildings (Sten Minör)
- Open innovation in software engineering using OSS: when to conceal and when to reveal? (Hussan Munir)
- Open Hack - Engineers without borders (Madelene Serenhov)
- Show and Tell: Why build Lego on your own if you can program your YuMi to it in 30 minutes?
- Software Research IRL - Reports from 8 years of EASE (Embedded Applications Software Engineering)
- TagOn - Automated analysis of patient flow and equipment location (Sten Minör)
10.00 Morning session
Keynote: Eric Giertz, KTH: Research and business development in Sweden
There has been strong links between engineering sciences and business development in Sweden during many decades but the interplay between academia and industry has changed considerably over time. In the complex society of today there are many different paths from excellent research to business development and commercial success. [Biography] [ presentation]
Robert Lagerstedt, Bosch: Mobile speed in automotive
Automotive industry is going through similar changes as the mobile industry has done, with increased usage of software, increased need of connectivity including software updates, and increasing need of speed in development. However using the best practices of software development from mobile industry does not always work in a good way due to different constraints and requirements. [ presentation]
Ulf Hagberg, ABB: Fifteen years of collaboration on domain-specific languages
During the last fifteen years, ABB in Malmö has collaborated with researchers at Lund University on the development of domain-specific programming languages for automation. This talk will discuss why collaboration on domain-specific languages is important for ABB, but also describe how the forms of cooperation have varied over time. [ presentation]
Björn Ekelund, Ericsson: The symbiosis between industry and academia
Mobile communication has up until now been mainly about connecting people. However, industry analysts are now forecasting that already within a few years a clear majority of the mobile connections will be between things, not people. This is very much what 5G is about.This, plus a constantly more complex value chain in ICT has made Ericsson take its strategy for industrial and academic collaboration to a new level. [ presentation]
Görel Hedin, Lund University: Lund Involvement in WASP
Wallenberg Autonomous Systems Program (WASP) is Sweden’s largest ever individual research program. The program addresses research on autonomous systems acting in collaboration with humans, adapting to their environment through sensors, information and knowledge, and forming intelligent systems-of-systems. Software is the main enabler in autonomous systems, and is an integrated research theme of the program. This talk describe how Lund University is involved in the program and how regional industry may utilize this involvement. [ presentation]
12.10 Lunch, demos & posters
12:50 LUFOSS Scholarship Ceremony
Award presenters Hampus Jakobsson and Carolina Sartorius. [ presentation]
13.20 Afternoon session
I fired my hardware architect! Who's going to write the program? (Flavius Gruian)
Low power, high performance, flexibility are today requirements for a large number of systems, often tackled by employing custom processor architectures, specialized for the intended application domain. Due to the exotic features of these processors (vector units, specific memory access), programming them is a challenging task, usually carried out close to the machine by experienced engineers. Building compilers that produce efficient code is a challenge, which we address in this project as part of HiPEC. [ presentation]
SCALARE - A framework helping industry to scale up software (Martin Höst)
In order to stay competitive, companies in all lines of business need to master software development. The SCALARE (SCALing sofwARE) project has developed a software Scaling Management Framework that will help industry scaling up software. In the presentation we will briefly describe the main results of the project, but also discuss experiences of collaborating in this international multi-partner project.
End-User Composition of Graphical User Interfaces for PalCom Systems (Björn A Johnsson, LTH)
In ubiquitous computing, end-user composition allows users to combine multiple single-purpose devices into new, interesting constellations. In PalCom — a ubiquitous middleware — this is achieved without the need to write program code. This talk present a solution that in the same way allows users to create Graphical User Interfaces (GUIs) for such systems without coding. The approach is to focus on presenting functionality in a GUI rather than attaching functionality to manually added components. When compared to another common tool, the presented editor exhibited roughly 10x shorter development times. [ presentation]
TagOn - Mitigating bottlenecks in emergency ward trough realtime tracking of equipment and people (Sten Minör, MAPCI)
MAPCI has, in collaboration with Sony Mobile, KnowIT, SP, and Skåne University Hospital, developed a system for wireless positioning of equipment, staff, and patients. The purpose with the project is to develop an IoT system to automatically measure, follow-up, and analyse patient flows in real time, locate vital equipment and staff in a full-scale pilot at the emergency room. This will give Region Skåne means to increase flow efficiency and utilization of resources, further to decrease waiting time, optimize value creating time for patients and staff, and optimize reception and care.
Local Positioning Systems (Kalle Åström)
Within the positioning laboratory at MAPCI, researchers from the centre for mathematical sciences, the department of automatic control, the department of EIT, Sony, Combain and Bitcraze, we collaborate on research questions concerning local positioning systems. These systems use sensor data, e.g. images, audio or radio, and new algorithms in order to make local maps of the infrastructure. Future sensor data can then be used to position the sensor relative to the map.
Using control theory to improve disaster recovery-aware backup strategies (Anders Robertsson)
Implementing geo-replication of data in data centers is expensive as every data update operation in the original data center must be replicated in the backup. Running the application and the replication service in parallel is cost effective but creates a trade-off between potential replication consistency and data loss and reduced application performance due to network resource contention. We model this trade-off and provide a control-theoretical solution based on Model Predictive Control (MPC) to dynamically allocate network bandwidth to accommodate the objectives of both replication and application data streams. [ presentation]
IoTaP – Internet of Things and People (Paul Davidsson, MAH)
IoTaP is a research center at Malmö University that focuses making the Internet of Things useful and usable. The main research areas are interaction technology, embedded intelligence and user-centered development. The research is conducted in collaboration with industrial partners and other organizations, and addresses important societal challenges in areas such as health, learning, energy and transport. [ presentation]
Exploratory Testing - Black or White? Using Focus Groups to Enhance Collaboration (Peter Visuri, Sony and Elizabeth Bjarnason, LTH)
Good and fruitful collaboration can be achieved when industry needs and research interests connect. In the EASE Theme E project industry and academia meet around common interests in decision support for testing. This has resulted in a successful workshop set-up around exploratory testing and the use of test charters. A topic that can address current industrial challenges within testing, while also being a promising field for novel research results. We have organized a focus group (workshop) where testers from Sony Mobile are provided with input on Exploratory testing and given the opportunity to try out different test charters. This provides the researchers with valuable empirical data, while the industry participants gain new insights, and time to reflect on how to improve on their testing practices. A win-win situation! [ presentation]
15:00 Coffee, posters & demos
15:30 Keynote: Mark van den Brand, Eindhoven University of Technology: Model Driven Engineering in High Tech Industry
Domain specific languages (DSLs) are becoming more popular, not only in academia but also in industry. The development of DSLs accelerated with the introduction of the Eclipse Modeling Framework (EMF). Existing tools, like openArchitectureWare, a collection of transformation tools for text-to-model (Xtext), model-to-model (Xtend) and model-to-text (Xpand), were integrated into EMF, new initiatives for transformation tools where launched, e.g. ATL, ETL, QVTo.
DSLs offer an increase in abstraction level. This contributes to an increase of the quality of the developed models as well as a reduction of development time. The increase in quality can be achieved via mappings to, for instance, model checkers to check properties. The development time is reduced by means of code generators.
There are a number of pitfalls when developing DSLs. The lack of proper modularity mechanisms of meta-models hinders reuse. Semantics is often encoded in the code generators; this hinders the development of verifiable model transformations. Meta-models appear to be less stable than grammars, which creates the problem of co-evolution. [ presentation]