Middleware for Robotics in Assisted living: A case study
Middleware for Robotics in Assisted living (Download)
Unicycle Steering by Brakes: a Passive Guidance Support for an Assistive Cart
Unicycle Steering by Brakes (Download)
Motion Planning in Crowds using Statistical Model Checking to Enhance the Social Force ModelMotion Planning (Download)
Behavioural Templates Improve Robot Motion Planning with Social Force Model in Human Environments
Behavioural Templates (Download)
Head pose estimation on depth data based on Particle Swarm Optimization
Head pose estimation (Download)
DALi Brochure
Want to know more about DALi, check out our brochure.
Brochure (Download)
Dali Team
Check how the work of each of our partners make this possible
We are eight full-time partners from the six European countries Italy, Spain, Greece, United Kingdom, France, and Austria. We joint our work experience towards the same goal "extending the people autonomous life beyond the home."
Indoor-Localization
Localization of single camera images in a relation to a previously created 3D model
The 3D model has been generated from a multitude of overlapping images resulting in a 3D point cloud representation by means of so called “Structure from Motion”. Single images to be localized did not contribute to model generation. Initially we see the 3D point cloud model and the pyramids represent the camera position of various still images that have been localized with respect to the model. In the fly-through sequence we move from camera to camera position and as we approach the perspective of the camera we can see how well the camera image fits with the underlying 3D point cloud data.
Robot path planning exploiting SMC and Social Force Model
Simulation of a robot moving in an environment populated by agents and obstacles. The direction of the robot is controlled by PLASMA, a statistical model checker (SMC), that avoids unsafe routes in the environment. The Social Force Model is used by the robot to predict the future positions of the agents along a convenient time horizon (2 seconds in these simulations). The magenta dashed line represents the path the robot has to follow and the circles represent the physical occupation of the agents.
Sensing Technologies
Head pose estimation on depth data based on Particle Swarm Optimization
We propose a method for human head pose estimation based on images acquired by a depth camera. During an initialization phase, a reference depth image of a human subject is obtained. At run time, the method searches the 6-dimensional pose space to find a pose from which the head appears identical to the reference view. This search is formulated as an optimization problem whose objective function quantifies the discrepancy of the depth measurements between the hypothesized views to the reference view. The method is demonstrated in several data sets including ones with known ground truth and comparatively evaluated with respect to state of the art methods. The obtained experimental results show that the proposed method outperforms existing methods in accuracy and tolerance to occlusions. Additionally, compared to the state of the art, it handles head pose estimation in a wider range of head poses.
The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° ICT-2011-288917.
The contents of this website reflect only the authors’ views and the European Union is not liable for any use that may be made of the information contained therein.