PercEvite · Project

Collision-Avoidance Eyes and Ears for Small Drones in Crowded Airspace

transportPilotedTRL 6

Imagine a small drone that can see, hear, and talk to everything around it — other drones, birds, buildings, power lines — and dodge them all on its own. Right now most small drones can only spot obstacles about 30 meters ahead, which is like driving with foggy glasses. This team built a lightweight sensor-and-radio package that lets drones detect obstacles much further out by teaching cameras to recognize danger the way our own eyes learn. They also gave drones a shared radio channel so they can warn each other, cutting the chance of a mid-air collision to nearly one in a billion.

By the numbers

10⁻⁹

Collision probability in collaborative avoidance scenario

10⁻⁶

Collision probability in non-collaborative avoidance scenario

~30 m

Current obstacle detection range the project aims to surpass

Consortium partners across 3 countries

Total project deliverables produced

The business problem

What needed solving

Small commercial drones today are effectively blind beyond about 30 meters and cannot reliably detect or avoid other aircraft, buildings, or obstacles on their own. This makes it dangerous and often illegal to fly drones beyond visual line of sight — the exact capability that delivery, inspection, and agricultural drone businesses need to scale. Without autonomous sense-and-avoid, every drone flight needs a human pilot watching, which kills the economics of fleet operations.

The solution

What was built

The team built and demonstrated a complete lightweight sensor, communication, and processing package for small drones. The final system (Deliverable 6.2) showed multiple drones autonomously avoiding each other and static obstacles in semi-outdoor conditions, using fused stereo vision, audio, and multi-radio communication (ADSB, 4/5G, WiFi) — 11 deliverables in total.

Audience

Who needs this

Drone delivery operators scaling beyond-visual-line-of-sight flightsUtility and infrastructure inspection companies using drone fleets near structuresAgricultural drone service providers operating multiple drones simultaneouslyUrban air mobility and air taxi developers needing certified collision avoidanceDrone manufacturers looking to integrate sense-and-avoid as a standard feature

Business applications

Who can put this to work

Drone delivery and logistics

SME

Target: Drone delivery operators and urban air mobility companies

If you are a drone delivery operator dealing with the risk of mid-air collisions as you scale to dozens of flights per day — this project developed a lightweight sensor and communication suite that enables autonomous obstacle avoidance with a collision probability as low as 10⁻⁹ in cooperative airspace. The system combines stereo vision, audio, and multi-radio communication (ADSB, 4/5G, WiFi) without heavy laser sensors, preserving payload capacity for parcels.

Infrastructure inspection

mid-size

Target: Utility companies and inspection service providers using drone fleets

If you are an infrastructure inspection company flying drones near power lines, bridges, or wind turbines and worried about crashes with structures or other aircraft — this project built a self-supervised perception system that detects ground-based obstacles far beyond the current 30 m range using only cameras and microphones. The demonstration showed multiple drones safely avoiding each other and static obstacles in semi-outdoor conditions.

Precision agriculture

any

Target: Agricultural drone service companies and large farm operators

If you are an agricultural drone service provider flying multiple crop-spraying or mapping drones over the same field — this project created an interoperable multi-drone coordination system using ADSB, 4/5G, and WiFi so your fleet can share position, speed, and waypoints in real time. The energy-efficient design means lighter sensors and more room for spray tanks or imaging payloads on each drone.

Frequently asked

Quick answers

What would it cost to integrate this sense-and-avoid system into our existing drone fleet?

The project data does not include pricing or per-unit cost figures. The system was designed to be lightweight and energy-efficient specifically to keep commercial viability in mind. Contact the consortium for licensing or integration pricing discussions.

Can this scale to manage dozens or hundreds of drones in the same airspace?

The collaborative avoidance system was designed for very densely used airspace, achieving a collision probability of 10⁻⁹ by exchanging position, speed, and future waypoints over ADSB, 4/5G, and WiFi. The demonstration (Deliverable 6.2) showed multiple drones avoiding each other and static obstacles simultaneously.

Who owns the IP and can we license this technology?

The consortium is led by TU Delft with 4 partners across 3 countries, including 2 industrial partners and 1 SME. IP ownership likely follows the standard EU grant agreement split among partners. Based on available project data, licensing terms would need to be negotiated directly with the consortium.

Does this meet aviation safety regulations like EASA requirements for drone operations?

The project was funded under SESAR (Single European Sky ATM Research), which is the EU program for modernizing air traffic management. The collision probabilities targeted — 10⁻⁹ for collaborative and 10⁻⁶ for non-collaborative scenarios — align with aviation safety standards. However, formal EASA certification status is not confirmed in the project data.

How mature is this — lab demo or flight-tested?

Deliverable 6.2 is described as a demonstration of the final developed system where multiple drones avoid each other and static obstacles in a semi-outdoor scenario. This indicates real-world flight testing beyond the laboratory stage, though not yet full commercial deployment.

Can we plug this into drones we already operate, or does it require new hardware?

The system combines stereo cameras, microphones, and multi-protocol communication hardware (ADSB, 4/5G, WiFi) into a lightweight package designed to maximize payload capacity. Based on available project data, integration with existing platforms would depend on weight and power budgets of your current drones.

Consortium

Who built it

The PercEvite consortium is compact and well-balanced: 4 partners from 3 countries (Netherlands, Belgium, France) with an even 50/50 split between academia and industry. TU Delft leads — a globally recognized university in aerospace and drone research. The presence of 2 industrial partners (including 1 SME) and the project's SESAR funding indicate strong alignment with European aviation industry needs. For a business looking to adopt this technology, the small consortium means fewer parties to negotiate with, and the industrial partners likely have practical experience bringing drone systems to market.

TECHNISCHE UNIVERSITEIT DELFTCoordinator · NL
KATHOLIEKE UNIVERSITEIT LEUVENparticipant · BE

How to reach the team

TU Delft, Netherlands — reach out to the Faculty of Aerospace Engineering drone research group

Order a report on this consortium See TECHNISCHE UNIVERSITEIT DELFT profile →

Next steps

Talk to the team behind this work.

Want an introduction to the PercEvite team? SciTransfer can connect you with the right people and provide a detailed technology brief tailored to your use case.

Order a report on this topic More in Transport & Mobility