Mercedes-Benz Digital Challenge, India

352 Registered Allowed team size: 2 - 5

This campaign is over.

idea phase
starts on:
Nov 23, 2019, 06:30 AM
ends on:
Dec 27, 2019, 12:25 PM
starts on:
Jan 01, 2020, 06:30 AM
ends on:
Jan 12, 2020, 02:30 PM


With our invention of the automobile, we moved the world for the first time and since then have relentlessly followed our passion to re-invent mobility!

After a series of successful student engagement projects and hackathons, Mercedes-Benz continues to stay invested in exclusive formats for enabling young talents in the academia to contribute with their innovative ideas and amazing technical solutions to fulfill real-world mobility demands.

Mercedes-Benz Digital Challenge, India - 2019 is yet another initiative to embrace the spirit of Innovation in the student fraternity across India. This challenge is jointly organized by Mercedes-Benz India, Pune and Mercedes-Benz Research & Development India, Bangalore and supported by DigitialLife@Daimler, Germany

This challenge is unique in more ways than one.

enter image description here

Breaking the 24-hour stereotype, budding engineers will have 6 weeks to ideate, provide their solution approach and implement those ideas. This will ensure the high quality of the deliverables.

Further, students do not have only a few minutes to “pitch” their idea. Breaking the second stereotype, the finalist teams will be invited to Mercedes-Benz Res. & Dev. India (MBRDI), Bangalore campus and have a full day to demo their solution, interact with experts and get suggestions from them.

In addition to the lucrative cash awards, chosen few finalist individuals to stand a chance to take part and present their solution in a grand tech symposium in Europe or visit the epicenter of automotive development in Germany - Stuttgart with a fully sponsored trip by DigitialLife Campus from Daimler. Further, the top teams will have a chance to be part of Auto Expo 2020 in New Delhi. Not to forget, if we find any of the ideas patentable and relevant, we will file the patent with the involved students as the co-inventors.

This challenge will provide a unique prospect to budding engineers of the country to co-create the future of mobility together with Mercedes-Benz and Daimler.

We invite you to make the most it and build the future of mobility together!

Eligibility:- Any student pursuing graduation from any degree/stream of IITs and a few select top colleges in India.


As inventors of the automobile, the passion to create a better future is part of our DNA. This involves not only the automobile but also the whole ecosystem around. Looking at the future of mobility with a lot more connected, electric and autonomous vehicles around, we have chosen the following challenge areas (grouped into themes), for you, the young minds, to deliberate, to ponder over and provide us a solution approach that you can demonstrate.

Till 26-Dec, the teams can register and submit their solution approach. Experts from Mercedes-Benz will be going through the submitted approaches on a daily basis and provide their decision on moving them to the implementation phase or not on the website on a regular basis. So earlier you get your solution approach shortlisted, the longer you get for the implementation!!

During the implementation phase, the teams will be provided with guidance through mentors from Mercedes-Benz. The code must be uploaded on the site (along with steps to build and deploy) and the solution deployed on any public cloud infra so that the results can be checked and verified by the experts. The deadline to finish the above is on 10-Jan. On 12-Jan, the finalists will be announced. They will be invited to MBRDI, Bangalore to demo and explain their solution to the jury. The results will be announced on the same day.

Steps to participate:

  1. You will need to download the solution submission template from here.
  2. Click on the Submit Idea button on the top right-hand side corner of the page.
  3. You will be expected to fill the same and upload it via the upload option while submitting the Idea.
  4. Choose your theme corresponding to your solution from the dropdown.
  5. Select Publish your idea.
AI based computer Vision
Deep Learning based driver monitoring system (activity & object recognition)

1. Deep Learning based driver monitoring system (activity & object recognition)
In recent years there has been a lot of focus on developing driver monitoring software for integration in passenger cars and other vehicles to facilitate better safety and other functions that improve the user experience. By studying a person’s posture and body movements, intelligent interior vehicle algorithms can draw conclusions about a person’s alertness, attention and focus.

Tomorrow’s cabin sensing features will include detection of passenger position, safety belt status and forgotten objects, as well as enabling multimodal functionality such as deeper AI and mood recognition.So the car is able to seamlessly transfer control of the vehicle to an awake and able driver, call for help in a medical emergency, or offer to play the perfect song for the moment.

Can you think of an interesting use case(s) in this area and implement them?

Customer engagement
Nexgen Car upgrade and finance option recommendation engine
Benz for kids!

1. Nexgen Car upgrade and finance option recommendation engine
Recommend the right car model and financing option to an existing car owner in India on or before their financing contract matures.

The objective of the challenge is to figure out the right offer for an existing car customer who has financed his/her car using a financing option so that they upgrade to a newer model. There are two components to making a great offer; the right car model and a timely financing offer. Through data analysis, make an offer at the right time so that customer feels compelled to upgrade to a new one. Factors influencing the model will be current contract status, term, maturity date, credit status, monthly payment, outstanding loan amount, current price of a new car (of different models) and current contract options.

2. Benz for kids!
Kids with their intrinsic energy and behaviour have to be kept engaged in the rear seats during commutes. Many a times parents do that by handing out their mobile phones or other hand-held devices which is certainly not their favourite or first preference. We would be interested to see solutions in this space where kids can get byte sized learning modules with gamification elements to make it intriguing. With several car and geo location data points from the car, there can be an array of opportunities and ideas which feed into entertainment for kids (and adults too).

Mobility Infrastructure
Safe Route & best way forward
Transport solution/algorithm for Indian urban needs
Mobility for Specially abled

1. Safe Route & best way forward
How many times have we taken not the shortest route as shown by the navigation because the shortest path has some desolate, dark stretches we want to avoid. Wouldn’t it be nice to have this information provided too? Most navigation solutions have options like shortest distance, fastest route, toll-free and other variants. With urbanisation, both in developed and developing countries and constantly evolving cities and suburbs there are extensive dynamic situations which influence our commute. Introduction of the ‘safest route’ option in navigation based on different factors, e.g., road repairs, accident and SOS data, newsfeed on disruption, based on reviews, (or anything else you can think of) can be a worthwhile option for the users.

2. Transport solution/algorithm for Indian urban needs (e.g. multimodal transport)
India’s transport demand has grown by almost 8 times since 1980 - more than any other Asian economy with thriving auto industry and allied economic growth. However, with this growth comes few health and welfare related challenges which needs to be addressed. One clear imperative for the country’s mobility paradigm is to build safe, clean, convenient and congestion free transportation. We are seeking ideas to optimize travel footprint by reducing congestion caused by passengers and goods flow in the urban context. Data-based models and measures such as intelligent routing and recommendation based on public infrastructure data, smart algorithms that make multi-modal and balanced mobility a more lucrative option than current.

3. Mobility for Specially abled
Mercedes-Benz is devoted to ensure that people with restricted mobility are easily and independently able to take to the roads, or be safely and comfortably carried as passengers. Inclusion of physically disabled, senior citizens, visually impaired, cerebral palsy and others for usage of all modes of mobility. We are looking for digital solutions with which our special consumers can leverage the currently available ecosystem in a better way.

Tech 4 Autonomous
Recognition of thin cross section objects changing direction
Image understanding for scene labeling based on sensor input

1. Recognition of thin cross section objects changing direction (e.g. a cyclist)
Object detection in self-driving cars is one of the most challenging and important impediments to full autonomy. Self-driving cars need to be able to detect cars and pedestrians to safely navigate their environment. In recent years, state-of-the-art deep learning approaches such as Convolutional Neural Networks (CNNs) have enabled great advances in using camera imagery to detect and classify objects. Recognizing thin cross section objects, which is also continuously changing direction (e.g. a cyclist) is even more challenging. Can you come out with your approach to do the same and demonstrate it?

2. Image understanding for scene labeling based on sensor input
For autonomous driving, vehicles typically have multiple sensors of complementary modalities such as cameras, LiDAR and RADAR. They generate a comprehensive and robust representation of the surrounding. Each sensor modality leverages its specific strengths to extract as much information as possible from the observed scene. Based on the combined sensor data, a detailed environment model is created. This environment model provides the basis for high-level tasks such as object tracking, situation analysis and path planning.

In order to perform these high-level tasks well, it is essential for an autonomous vehicle to not only distinguish between generic obstacles and free-space, but also to obtain a deeper semantic understanding of its surroundings. Detailed semantic information of similar quality has to be extracted independently from each of the sensor modalities to maximize system performance, availability and safety.

Can you come out with your approach which is different from existing ones and can provide better results?

View All Notifications