New FFG project granted

In this project, we are tackling the challenges of high-precision state estimation with the application to approach, landing, and subsequent take-off of a multi-copter on a landing pad for autonomous recharging.The goal is to develop a reliable algorithm for state estimation using multiple sensor modalities in a consistent and real-time multi-sensor fusion framework which will serve as back-bone for the high-precision maneuvers required during take-off and landing for autonomous recharging. The focus on estimator consistency and real-time is particularly important to guarantee the correct signal weighting of each sensor yielding an overall robust state estimate on-board the system. In addition to the data fusion of multiple sensors in a given sensor suite, we also investigate the question of how to increase modularity of such a framework such that sensor signals and entire sensor modules can adaptively be added and removed in-flight without disabling state estimation for reliable UAV navigation. A typical example is when a drone enters GPS shadowed areas or where GPS suffers severely from multi-path issues.


We have been awarded a research grant in order to tackle the problems from our project proposal: AnyView3D. In AnyView3D, we want to meet the challenges of 3D reconstruction and semantic scene classification using multiple cameras. Our approach is based on a SVS(Surround View System) Setup and uses at least four wide-angle cameras around a vehicle. The aim of the project is the automatic generation of a complete environment model of the short-range vehicle  surroundings. The key issue is how classic 3D image processing techniques using a plurality of cameras can be integrated into a unified architecture, so that hardware and software redundancy are used optimally.

IVISO Stereo Camera

We are happy to annouce our custom build IVISO Stereo camera. It consists of 3 cameras. Two greyscale cameras with a resolution of 1280x960 connected via USB2 and a baseline of 12.5 cm. In between we have setup a RGB camera with the same resolution to enrich reconstructed 3D point clouds with colorfull textures. On a standard laptop we currently achieve 12fps for the computation of depth maps with a target resolution of  640x480 . The matching is done using a Census transform in a multi-scale fashion. Next, we will test our stereo visual odometry algorithm.

Page 1 of 2



Find out about IVISO, its services and its attitude towards R&D.



Find out about our areas of expertise and the diverse technologies and domains we master.



You seek a challenging job opportunity at the cutting edge of technology. Please see our open positions.

EMail:     This email address is being protected from spambots. You need JavaScript enabled to view it.

Tel.: +43 660 66 31 877

VAT: ATU69357117

Trade register: 438366g

IBAN: AT90 3427 7000 0212 4667


Legal Form: Limited Liability Company 

@ Iviso 2015