Start: Sunday - October 27th, 2019.
Deadline: Sunday - November 10th, 2019.
It takes 10 to 15 minutes!
- Personal Data (CV upload only in pdf. The file name must have the ending .pdf - no capital letters)
- Study-related Data
- Skills & Experience (drop-down menu)
- Programming language you are proficient in (drop-down menu)
- Choose: Apply individually or as Team members (drop-down menu)
- Discription of the ten challenges (read ten texts attentively)
- Your top 5-challenge choices (drop-down menu)
- Organizational matters (drop-down menu)
- Declaration of Consent (read attentively and activate two "check-boxes")
- Accuracy of Data (activate one "check-boxes")
- Verification (enter the verification code)
- Press button: SEND APPLICATION
Students of diverse fields from all the universities in Munich will be able to participate in this year’s Science Hack. They will display their creativity, passion and problem-solving skills towards solving challenges provided by exceptional chairs of TUM. The challenges will be thought provoking and highly topical and based on making the world greener, healthier and sustainable for the future.
We thank our Sponsors & partners!
ALTAIR engineering | BMW group | HUBERT BURDA media | ITK engineering | PIXIDA gmbh | SIEMENS AG | WACKER chemie
- TUM Chair for DATA PROCESSING . Prof. Dr.-Ing Klaus Diepold
- TUM Assistent Professorship of ENVIRONMENTAL SENSING AND MODELING . Prof. Dr.-Ing. Jia Chen
- Kick-Off evening on November 26, Tuesday (6-9 pm) - mandatory
- Science Hack weekend on December 7th & 8th
- Kick-Off evening: Immatrikulationshalle, TUM campus Munich
- Science Hack weekend: AudiMax, TUM campus Munich
- 1st: 1.000,00 € / team
- 2nd: 600,00 € / team
- 3rd: 400,00 € / team
Eighteen of the Nineteen warmest years have occurred since 2001, except for 1998. The year 2016 ranks as the warmest on record. Whether is due to human activity or natural variability, it’s well known that the C02 concentration in the air plays a big role in making our world warmer.
But where C02 comes from? Why is it increasing? Can we build models to quantify its average concentration?
Let’s build a dynamic model which taking into accounts different factors can predict the trend of C02 concentration. For doing it, you can use any mathematical formulation you like, and you think is more suitable (linear regressions, look up tables, finite difference equations, machine learning). The only restriction is that the model needs to run on our Mathematical Environment Altair Compose.
We are not expecting that you develop an extremely accurate model. Instead we want you to develop a simple and interactive model which can predict the C02 concentration once the user defines some inputs.
For this reason, we suggest that you pick up a limited number of factors (such us deforestation, fires, C02 sinks, industrial growth…) and try to think at smart ways to model their effects on C02. For example you can grab historical data to build linear regression models or you can make predictions on factors (e.g. percentage of C02 absorbed by forests,…), or you can leverage equations provided by literature.
We suggest that you divide this challenge in four steps:
- Define the factors you want to include in the model;
- Gather the whole data you need to define the initial state of the system and its evolution;
- Divide into groups to model each factor. In this phase each factor shall be modeled as an independent function (at least as much a possible) to make easier the final phase;
- Assembling all the scripts together and generate all the reports/plots;
- Validate the model with historical data;
The final purpose of this challenge is to have a runnable model. A common user should be able to run this model and, defining a certain number of inputs, he should obtain the concentration of C02 for a certain number of years in the future and an estimate of global warming.
And when you will have a runnable model, we would like you to think at:
“So which scenario (combination of factors/players) should we aim at to maintain the C02 concentration in the air below a certain level?”
Sustainable use of resources is getting one of societies’ top challenges. The technical advancements, especially in the field of information technology, within the last decade enable completely new opportunities to manage these challenges. In the context of production, so called “Smart Wearables” are more and more used to assist the workforce in an intelligent manner in order to be more efficiently conducting their daily routines. In this segment, the rapid processing of information and analysis of data plays a significant role.
The Science Hack´s challenge focuses on making an IoT device which can effectively support a worker within automotive production in preventing false assembly of parts.
Makerspace will provide the bill of material data sets, hardware equipment to build a product and its information technology interface (e.g. Rasperry Pi) as well as automotive assembly knowledge. Based on the provided data set, the device gives feedback to its carrier.
- Hands-on skills to develop and make a prototype
- Apply practical programming knowledge, e.g. Visual Basic, to allow data interpretation
- Database knowledge and communication protocols i.e. WLAN, Bluetooth, RFID, …
- Engineer knowledge in the field of automotive production
- Futuristic mind and teamplayer to conceptualize ideas on how workforce can optimize the use of resources
- Present a prototype along with your interpretation of possible design alternatives
To allow the environment to recover from stress and pollution created by human kind, each German citizen should emit no more than 2 tons of CO2 per year– which is dramatically less than the current amount of CO2 produced by German individuals per year (8 – 10 tons). However, it is very difficult for individuals to estimate to which degree the consumption of groceries, clothing, transportation, etc. contributes to their personal carbon footprint. Hence, we would like you to develop an app or platform that helps to identify major contributors to the individual’s carbon footprint and suggests better alternatives respectively shows options to compensate for the associated greenhouse gas emissions.
- Tracking of the own carbon footprint
- Rating of the amount of CO2 emitted due to goods / activities (e.g. class 1: low emission, class n: very high emission)
- Suggestion of suitable options for carbon emission compensation (incl. implementation of associated payment processing)
Components like a microcontroller, a touch screen, photodiodes, casing and energy supply are combined to develop a handy universal remote control.
If the devices communicate via infrared, there is a plethora of possibilities:
- Recording of the original remote control’s signals
- Management of multiple devices, e.g. TV, Blu-ray player, beamer (+ canvas), etc. without searching for the correct remote control
- Realization of complex tasks by combining individual commands addressing several devices (e.g. lower beamer canvas, switch off radio, switch on beamer, switch on receiver, select TV channel)
The challenge does not solely consist of programming but also of designing an intuitive GUI to make the remote control’s possibilities easily accessible for the user.
On the one hand public debate on air pollution in German cities as well as in cities worldwide is ongoing - on the other hand ubiquitous mobility gained by owning a car has become a pillar in our society. As car exhausts contribute to air pollution, both car manufacturers and private customers could contribute to their reduction- the former by developing more efficient combustion engines and electromobility, the latter with eco-friendly driving.
This Science Hack’s challenge focuses on the development of a web app to promote eco-friendly driving by providing an eco score for every completed trip.
Pixida will provide a dataset of trips containing timestamps, raw and matched GPS coordinates, velocity and acceleration values. This data can be enriched with elevation profiles, speed limits, route types, etc. using e.g. the HERE API.
- Engineer features characterizing the ecofriendliness of a trip.
- Develop a basic web app to compute and visualize the eco-friendliness per trip.
- Present the web app along with your interpretation of eco-friendly driving
- Conceptualize ideas on how users can be stimulated to optimize their eco score.
The goal of this proposal is to develop a low-cost solution to monitor how people perform a task during a VR training. This solution will be realized using trackers that can be easily attached to the person’s body parts ( hand-feet). Up to 6 trackers will be used to recreate the people's movement in the virtual world using an avatar. The avatar should not only follow in real-time the movements of the user in the real world and replicate the pose of the body.
Develop a VR application that will include an avatar replicating real-time movements of a person wearing the tracking devices. Compare real-time movements with a predefined right body posture. The app gives user feedback if the deviation from right execution of the task is observed. As task picking up a box can be used
- Several training solutions in VR are already available
- Usually, when people learn to do something in one way, they repeat in the future in the same way
- None of these solutions takes into consideration the ergonomic part
- Solutions like suits equipped with sensors to monitor people‘s movement exist on the market but very expensive ( >5.000€)
- Other solutions include cameras that can recreate the movement of the body.
- However, such solutions are complicated to be realized because of the sensible data (personal information) and their handling
- Ergonomics are important because it correlates directly with a person‘s health
- Repeating an action in the wrong way has an impact on the person’s body
Siemens will provide:
- 1 Varjo VR glasses ( best VR glass-highest resolution on the market)
- 1 gaming laptop – 1 powerful workstation
The booming world population, urbanization and climate change creates a great pressure on global food production, forecasted to increase by over 50% by 2050. The ability of natural resources, to meet this intensifying demand, is meanwhile constrained. How technology can help us feeding the world is the challenge tackled by the project Food Lab of Siemens Corporate Technology. Thanks to our expertise in digital technology and robotics, a fully autonomous and sustainable farming system is created - one that is leaner, resource-efficient with high food quality of fresh foods. Now, it is time for transformative steps with the help of Augmented Reality (AR). Using the Siemens MindSphere open IoT operating system, students have the opportunity to combine vision and sensor data of the environment and plants to create an AR app that allows farmers to control and monitor the growth of any crop, from anywhere at any time. From seed to harvest, there will be a full traceability thanks to digital technology, growing food that the world can trust.
2 visualization modes:
- General overview of all boxes: Trigger is image of whole table with boxes
- Individualized overview of box: Trigger is (i) QR code or (ii) zoom on box
General overview of all boxes:
- View of air temperature, air humidity
- Control over LED light
Individualized overview of box:
- View of growing time, soil sensor data, beacon data, crop type
- View of history of soil sensor data
- Control over irrigation system, take a picture by camera of tablet and/or camera of robot
Whether for greenhouse gases, climate change or climate protection – one term plays a central role: carbon dioxide, or CO2 for short. Climate protection is an essential part of WACKER’s sustainability strategy!
The company has done a lot to lower its carbon footprint and has set goals to further reduce it. It checks on the effectiveness of its climate protection measures using a group-wide CO2 balance. But how can each employee contribute to the strategic “Footprint Down” target? Well, we’d like to include everyone in showing how.
So, we are looking for an easy-to-use and integrated CO2 calculator to specify the contribution WACKER employees make. For example, when going on a business trip, employees can immediately see how much CO2 they are causing or how much they could save (e.g. by taking the train instead of flying). Other parts of the workday, such as the impact of commuting methods, the consumption of food and drinks or the use of electric or electronic devices, etc., should be included.
The calculator should promote awareness of climate-friendly workplace routines and encourage taking action by suggesting suitable measures. Users can create scenarios and make comparisons to underline their final choices.
In addition, employees should have the opportunity to trigger WACKER internal projects on the reduction of emissions or carbon compensation. Therefore, the tool should be accompanied by a concept of how this emission reduction initiative or the compensation could work for WACKER.
The calculator should function on all devices (desktop, tablet and smartphone).
For movement and control you can use current machine learning and control techniques. Sensor technology and monitoring tasks can also be easily taken over by novice programmers. If you want less hardware, you can also take care of the user interface on your smartphone. For the handicraft enthusiasts, the complete design of the robot and the assembly using Robotis Stem Kits is ideal.
- Raspberry Pi
- Robotis Stem Kit
- Computer vision and light sensors
- Moisture and environment sensors
- Robot concept and construction
- App/webserver layout
- Communication Interface
- Control and learning algorithms
- Test cases
- Team Logo and presentation
The task is to develop a graphical user interface with which the global emission data of CO2 and CH4 can be vividly displayed. For this purpose we provide the data from different global emission inventories such as EDGAR, ODIAC, etc.
In the end, the user should be able to choose between the different inventories as well as to apply spatial and temporal restrictions. Figure 1 (below) shows an example of total global CO2 emissions.
The main deliverables of this challenge are as follows:
- An interactive time series of GHG emissions in which the user can define the area of interest and the time period (as in Figure 2)
- An interactive map (in .gif format) illustrating the spatial variation of the emissions of a user-defined area through time.
Recommended programming language is Python or R.
At Science Hack, you work for two days in teams of up to six people and with professional support on exciting and interesting challenges from the worlds of business and science.
Each problem combines issues from computer science with another scientific discipline.
The coolest projects are rewarded with attractive prizes!
While programming experience is almost essential for every team, the challanges will encourage you to use a diverse set of skills in approaching complex problems, so you don't have to be a coder to join.
Try to form teams that include people from different fields, but also make sure that there are coders in the team.
Yes, please bring a laptop. There will be WiFi and enough mains sockets for everyone.
Yes, only students and PhD-students can participate.
Not directly, but the application form has field for up to three other participants you'd like to see in your team.
Everyone has to apply on their own, but we try to take your wishes into account during the selection process.
We encourage you to form teams during the event. There will be many highly talented people from all the different fields so it is a great change to meet and learn from new people.
We, the coordination team for the 2. Science Hack will select the applicants.
Yes, there will be free food and drinks for the whole event duration.
We will provide camp beds in separate rooms so you can have a good rest at night (you don't have to, of course).
The first edition of TUM Science Hack hosted students from 7 different fields of studies and was supported by 5 partner companies. It was an incredible opportunity for students and companies to interact in finding unique solutions to important problems. Science Hack 2018