ChallengeUpdated on 16 January 2026
DB Regio GPS Challenge
Head of Startup Scouting at DB mindbox | Deutsche Bahn AG
Berlin, Germany
About
01 Vision
Precise vehicle positioning is the key to stable operational processes, higher automation, and a better overall system in regional rail transport. However, in current practice, positioning repeatedly reaches its limits—especially in areas where multiple tracks run close together or vehicles are parked.
With the GPS Challenge 2026, DB Regio invites external experts to find new, purely data-driven approaches to intelligently process existing position data in order to enable track-accurate positioning around the clock.
02 Background
In daily regional rail operations, existing solutions do not always allow for a clear determination of which track a train is on. The result:
-
Limited transparency for control centers and dispatching
-
Restricted automation of processes
-
Untapped potential in planning, operations, and maintenance
At the same time, traditional solutions such as vehicle retrofitting or additional hardware are not practical—they are costly, time-consuming, and cannot be scaled quickly.
The key question therefore is:
How can the quality of existing location data be significantly improved—purely by using data and algorithms?
03 Objective
We aim to introduce an innovative, scalable solution that:
-
Significantly increases the accuracy of vehicle positions from existing systems
-
Enables track-accurate positioning in relevant operational areas
-
Can be used in real time or near real time (processing speed in production: approx. 50 messages per second)
-
Works transparently and comprehensibly with access to source code (no black box)
-
Enables validation of output data according to FzO-BO criteria
The long-term goal is a nationwide, 24/7 available and operationally viable train positioning system for regional rail transport.
The GPS Challenge represents the first step and enables DB Regio to identify the best external team (i.e., a startup, technology company, or group of experts with founding intentions) with whom a 100-day Proof of Concept (PoC) will be developed and implemented after the challenge.
04 Selection Process
The DB Regio GPS Challenge is organized as a multi-stage process.
Applications can be submitted until 15 March 2026 via www.dbmindbox.com. All applications received by this date will be evaluated by an expert team from Deutsche Bahn.
For further information visit our Challenge Website:
https://www.dbmindbox.com/mindbox/current-calls/db-regio-gps-challenge
Topic
- Artificial Intelligence
- Machine Learning
- Big Data
- Edge/Cloud computing
Type
- Proof of concept/pilot testing
- Access to clients
- Acceleration/Incubation, mentoring and co-working spaces
Organisation
Similar opportunities
Challenge
Power Saving Solution for Personal Tracker and Logistics Tracking Platform
Wing Fai Stanley Ng
CEO at AI Cities Limited
Kowloon Bay, China (Hong Kong)
Challenge
Automotive In-Cabin Passenger Entertainment & Experience
- 5G
- Big Data
- VR/AR/MX
- Multimedia
- Audiovisual
- Acquisition
- Connectivity
- Voice / Audio
- Neurosciences
- Edge/Cloud computing
- Artificial Intelligence
- Emotion recognition software
- Proof of concept/pilot testing
- Cybersecurity/data security/cryptography
- Client-provider collaboration (commercial agreement)
Jens Kronen
Senior Technology Scouting & Partnership Manager at HARMAN Automotive
Munich, Germany
Challenge
Data-driven Grid Resilience: Planning & Asset Intelligence for Future-Proof Transmission Networks
- IOT
- Big Data
- Machine Learning
- Artificial Intelligence
- Proof of concept/pilot testing
- Systems Integration & Advanced Manufacturing
- Investment (Venture capital and corporate venturing)
Filippo Casagrande
Terna
Padova, Italy