All projects

Technology

Integrated LoRa Sensor Monitoring & Analytics System

Turning raw LoRa telemetry from tree-mounted sensors into live dashboards that answer watering and growth questions - built in one week.

1 week · From nothing to a working system

Ottronic E-Systems (TUM 1000+ Project Week) Data Pipeline & IoT Analytics
Geospatial Grafana InfluxDB IoT LoRa Node-RED Time-Series
LoRa Sensor Monitoring & Analytics

The challenge

Ottronic had fitted trees in Austria with two kinds of LoRa sensor to monitor tree health: a strain sensor measuring growth and contraction of the trunk, and a tensiometer measuring soil moisture. The nodes also reported ambient temperature and their own battery voltage.

The data was flowing, but it wasn't actionable. The LoRa packet carried no field identifying which sensor type a reading came from, so humidity and expansion measurements arrived interleaved in the same stream. Nothing was being retained for historical analysis, and the field team had no way to ask the data "which tree needs watering?" or "is this trunk growing, or shrinking from water stress?"

The approach

This work was carried out during a TUM 1000+ Project Week: an interdisciplinary team of four master's students spent a week on site with Ottronic in Fohnsdorf, Austria. We built the chain from sensor to dashboard: LoRa gateway → Node-RED → InfluxDB → Grafana.

  1. Deriving sensor type from the data. The packet didn't declare its type. We parsed the raw string (@;R69;S12;131;22960;173;24;!) and wrote a numerical rule on the fourth field: above 250 the reading is trunk expansion, at or below 250 it is soil humidity. Every packet is then routed to the right measurement with the right column names.
  2. Storage. Readings are kept as time series in InfluxDB 2.0, making historical analysis possible for the first time.
  3. The map. A Geomap dashboard in Grafana shows every node at its location, with soil icons for humidity sensors and tree icons for expansion sensors. Clicking a marker reveals the latest values and a link through to that sensor's detail dashboard.
  4. Eight automated insights. Alongside the raw graphs we added indicators that turn data straight into a maintenance decision: a four-level watering recommendation, drought risk, temperature stress (frost and heat alerts), drying rate over a rolling two-hour window, overwatering risk, battery status, a 100-point tree health score, and trunk diameter change rate - from "shrinking fast" to "growing fast".
  5. Adjustable by design. Every threshold can be changed from the dashboard, so different tree species or climates need no code changes.

By the end of the week the team handed over the working system plus a step-by-step guide so Ottronic can add and remove sensors themselves.

The outcome

Results that moved the needle.

1 week

From nothing to a working system

On site during a TUM 1000+ Project Week, with a team of four

8

Automated maintenance insights

From watering recommendations to battery status and trunk growth

2

Sensor types separated from one stream

The packet carried no type field; a numerical rule resolved it

Want similar results for your project?

Every project above started with a conversation. Let's figure out what yours needs.

Keep exploring

More projects.

Other · A rail infrastructure engineering firm

Rail catenary pole placement automation

Weeks of expert engineering work, reduced to seconds.

View
Healthcare · Decoto International

D-Risk - MedTech Marketplace with AI Company Profiling

A three-sided marketplace linking medtech startups with investors and specialist freelancers - matched through AI document profiling.

View
Other · A certified accountant

Revenue Administration MCP Server

An assistant that reads Turkish tax legislation from its official source at the moment you ask and answers with the article behind it - a system a certified acc

View
Technology · Fraunhofer IKS, Munich

Reliability of LLMs in Safety-Critical Requirements Engineering

A controlled experiment measuring what an ungrounded, off-the-shelf chatbot contributes to safety-critical engineering.

View
Technology · Fraunhofer IKS, Munich

DSGENAI - AI Safety Requirements Engineering Platform

Stabilising and modernising an AI-driven safety-requirements platform - Flask to Streamlit, GPT-5.1, and critical data-leak fixes.

View
Technology · Independent project

Retrieval-Augmented Generation (RAG) Documentation Assistant

A RAG assistant that turns stacks of PDFs into a searchable knowledge base where every answer traces back to the source document.

View
Technology · Technical University of Munich

LLM Prompt Optimization for Legal-Clause Classification

Comparative research that lifted F1 from 0.62 to 0.76 on Terms-of-Service clause classification through automatic prompt optimisation alone - without retraining

View
Technology · Siemens

Smart Contract Analysis with NLP

An NLP system learns from Siemens' legal team's past contract revisions, flags the same clauses in a new contract, and proposes the edit that was made before.

View
Other · Scoutium

Football Player Potential Prediction Model

A classification model predicting whether a player will be marked "highlighted" from 39 scout attribute scores - ROC-AUC 0.86 under 10-fold cross-validation.

View
Other · Scoutium

Football Player Ranking System

A scoring engine that ranks players not by total score, but by how many attributes they exceed the statistical average for their own position.

View
Other · Scoutium

Football Player Position Recommender System

A recommender that compares a player's attribute profile against the profiles of other positions and finds players could be more valuable in a different role.

View
Other · Turkish Airlines

Automated Connecting-Flight Optimization

A tool that recalculates every connection possible through the hub - day by day, with passenger volumes - when a single flight's time is shifted.

View
Other · Turkish Airlines

Flight Passenger-Count Prediction

A demand model predicting booked passengers on one-stop routes from a flight's own characteristics, selected by comparing eleven regression models.

View
Technology · Istanbul Bilgi University

Turkish Image-Captioning Benchmark on MS COCO 2014

A human-verified Turkish caption dataset covering all of MS COCO, plus five models trained on it - a new reference point for Turkish image captioning.

View
Finance & Fintech · CFA Poland (hackathon)

ESG Diversity & Sentiment Solution - CFA Poland Hackathon

An ESG prototype scoring gender diversity and news sentiment - 2nd place among 44 teams from 28 universities.

View
Other · Boğaziçi University IT Club (hackathon)

Automated ESG Scoring System - HackBogazici

An automated ESG scoring engine built from four scraped data sources - 2nd place among 14 hackathon teams.

View