Embedded Perception & Autonomous Systems

Computer Vision Laboratory · Department of Automatic Control and Robotics · Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering · AGH University of Krakow, Poland

Our research focuses on embedded perception systems that process, in real time and at low power, high-resolution video streams, LiDAR point clouds and data from event cameras – for autonomous mobile robots, including UAVs, UGVs and legged robots.

Embedded perception & control SoC FPGA acceleration Event-based perception Neuromorphic computing Autonomous mobile robotics Real-time edge AI
Our research Publications Theses for students
Tomasz Kryjak
Tomasz Kryjak, PhD
Leader of the Group
156 publications
195 supervised theses
9 research projects
10 partner countries

What we work on

Computer vision

4K video processing, object detection and tracking

Real-time pipelines for high-resolution video, combining classical vision algorithms with deep networks.

Event-based vision

Low-latency perception with event cameras

Perception with event cameras (DVS): filtering, visual odometry, tracking and detection from sparse, microsecond-latency event streams.

Neuromorphic computing

SNNs, event-based audio, neuromorphic control

Spiking Neural Networks, event-based audio processing and neuromorphic control for advanced mobile robotics.

Embedded AI

CNNs and GNNs for real-time, low-power inference

Deep neural networks, including Graph Neural Networks for sparse data, compressed and optimised for on-device deployment.

Hardware acceleration

FPGA, SoC FPGA and embedded GPUs

Zynq, Zynq UltraScale+, Versal, NVIDIA Jetson and emerging neuromorphic hardware.

Autonomous drones

perception and control (classical, MPC, RL)

Perception: detection, tracking and visual navigation. Control: classical, MPC, RL and hybrid controllers.

News

Conferences →

Featured publications

  1. TRETS
    SIRA: scaled-integer range analysis for optimizing FPGA dataflow neural network accelerators
    ACM Transactions on Reconfigurable Technology and Systems, 2026
  2. JSA
  3. ARC
  4. CVPRW
All publications →

Our team

Research & Teaching Staff
Marek Gorgoń Marek Gorgoń Professor Marcin Kowalczyk Marcin Kowalczyk Research & Teaching Assistant Krzysztof Błachut Krzysztof Błachut Research & Teaching Assistant Hubert Szolc Hubert Szolc Research & Teaching Assistant Mateusz Wąsala Mateusz Wąsala Research & Teaching Assistant Michał Daniłowicz Michał Daniłowicz Research & Teaching Assistant Piotr Wzorek Piotr Wzorek Research & Teaching Assistant Kamil Jeziorek Kamil Jeziorek Research & Teaching Assistant
PhD Students & Engineers
Konrad Lis Konrad Lis PhD Student Mateusz Gołąbek Mateusz Gołąbek PhD Student Błażej Szargut Błażej Szargut PhD Student Remigiusz Mietła Remigiusz Mietła Research Engineer

Partners & funding

Project funding Project funding Project funding Project funding Project funding Czech Technical University in Prague Paderborn University AMD Research & Advanced Development Norwegian University of Science and Technology (NTNU) CEA-List Grenoble Keio University Universidad de Sevilla National Institute of Applied Sciences (INSA) of Rennes NOVA School of Science and Technology, NOVA University Lisbon University of Southern Denmark (SDU) Polytechnique Montréal Sorbonne Université

AGH AVADER Student Research Group

We supervise AGH AVADER, a student research group with four sections – Drones, FPGA, Neuromorphic and Media. Its members build real robotic systems, take part in international competitions and often turn their projects into bachelor's and master's theses.

Join AVADER →