A5088910768- Robotic Locomotion and Control
- Prosthetics and Rehabilitation Robotics
- Modular Robots and Swarm Intelligence
- Robotic Path Planning Algorithms
- EEG and Brain-Computer Interfaces
- Stochastic Gradient Optimization Techniques
- Parallel Computing and Optimization Techniques
- Advanced Memory and Neural Computing
- Optical Imaging and Spectroscopy Techniques
- Control and Dynamics of Mobile Robots
- Biomimetic flight and propulsion mechanisms
- Neuroscience and Neural Engineering
- Robotics and Automated Systems
- Advanced Data Storage Technologies
- Cognitive and developmental aspects of mathematical skills
- Advanced Manufacturing and Logistics Optimization
- Quantum Computing Algorithms and Architecture
- Robot Manipulation and Learning
- Adversarial Robustness in Machine Learning
- Human Motion and Animation
- Biomedical and Engineering Education
- Educational Games and Gamification
- Child and Animal Learning Development
- Bluetooth and Wireless Communication Technologies
- Micro and Nano Robotics
Maastricht University
2015-2025
École Polytechnique Fédérale de Lausanne
2010-2018
École Polytechnique
2011
École Normale Supérieure - PSL
2010
University of Zurich
2008
ETH Zurich
2007-2008
Imagine a world in which our furniture moves around like legged robots, interacts with us, and changes shape function during the day according to needs. This is long term vision we have Roombots project. To work towards this dream, are developing modular robotic modules that rotational degrees of freedom for locomotion as well active connection mechanisms runtime reconfiguration. A piece furniture, e.g. stool, will thus be composed several activate their joints together implement locomotor...
Bioinspired legged locomotion comprises different aspects, such as (i) benefiting from reduced complexity control approaches observed in humans/animals, (ii) combining embodiment with the controllers and (iii) reflecting neural mechanisms. One of most important lessons learned nature is significant role compliance simplifying control, enhancing energy efficiency robustness against perturbations for locomotion. In this research, we investigate how body morphology combination actuator design...
We present Oncilla robot, a novel mobile, quadruped legged locomotion machine. This large-cat sized, 5.1 robot is one of kind recent, bioinspired class designed with the capability model-free control. Animal in rough terrain clearly shaped by sensor feedback systems. Results show that agile and versatile possible without sensory signals to some extend, tracking becomes robust when control added (Ajaoolleian 2015). By incorporating mechanical blueprints inspired from animals, observing...
This study explores the relationship between children's academic performance and their results on cognitive function tests. Traditionally, test scores often reflect overall accuracy or speed. Yet, these are influenced by both diverse executive functions (EF) non-EF processes. To isolate specific functions, alternative scoring methods have been developed, which aim to measure one more purely. We investigated whether combining traditional with claiming strategy use improves prediction of...
Augmented reality (AR) enhances the user's environment by projecting virtual objects into real world in real-time. Brain-computer interfaces (BCIs) are systems that enable users to control external devices with their brain signals. BCIs can exploit AR technology interact physical and explore new ways of displaying feedback. This is important for perceive regulate activity or shape communication intentions while operating world. In this study, twelve healthy participants were introduced asked...
We present a floating-gate synaptic circuit that updates its weight according to the Spike-Timing-Dependent Plasticity (STDP) rule. The (or voltage) is updated only if time difference between pre- and post-synaptic spikes falls within learning window. update implemented through tunneling injection mechanisms which can be tuned for very long constants up seconds. novelty of this are turned on when correlation pre postsynaptic activity significant. additional benefit non-volatile technology...
This work addresses the instability in asynchronous data parallel optimization. It does so by introducing a novel distributed optimizer which is able to efficiently optimize centralized model under communication constraints. The achieves this pushing normalized sequence of first-order gradients parameter server. implies that magnitude worker delta smaller compared an accumulated gradient, and provides better direction towards minimum gradients, turn also forces possible implicit momentum...
We present a Bluetooth scatternet protocol (SNP) that provides the user with serial link to all connected members in transparent wireless (BT) network. By using only local decision making we can reduce overhead of our dramatically. show how SNP software layer simplifies variety tasks like synchronization central pattern generator controllers for actuators, collecting sensory data and building modular robot structures. The whole BT stack including new is implemented on single memory chip. To...
Roombots (RB) are self-reconfigurable modular robots designed to study robotic reconfiguration on a structured grid and adaptive locomotion off grid. One of the main goals this platform is create furniture inside living spaces such as homes or offices. To ease control RB modules in these environments, we propose novel more natural way interaction with grid, called Natural User Interface. In our method, user commands using pointing gestures. The user's body tracked multiple Kinects. also...
Abstract The correct and robust recognition of traffic signs is indispensable to self-driving vehicles driver-assistant systems. In this work, we propose evaluate two network architectures for multi-expert decision systems that test on a challenging Traffic Sign Recognition Benchmark dataset. implement individual experts in the form deep convolutional neural networks (CNNs). A gating CNN acts as final unit learns which expert CNNs are likely contribute an overall meaningful classification...
This study aims at providing a control-learning framework capable of generating optimal locomotion patterns for the modular robots. The key ideas are firstly to provide generic control structure that can be well-adapted different morphologies and secondly exploit coevolve both morphology aspects. A combining robot morphology, environment on top them optimization evolutionary algorithms presented. details components some preliminary results discussed. (C) Selection peer-review under...
Brain-Computer Interfaces (BCIs) are playing an increasingly important role in a broad spectrum of applications health, industry, education, and entertainment. We present novel, mobile non-invasive BCI for advanced robot control that is based on brain imaging method known as functional near-infrared spectroscopy (fNIRS). This the concept automated autonomous intention execution (AutInEx), is, possibly very complex actions action sequences intended by human through robot.
Assessment tests are crucial but often viewed as tedious work by the participants and experimentalists. Consequently, research has been done to integrate assessments into serious games automate observations using computer vision or wearable sensors. However, use of camera generates privacy concerns. Moreover, behavior a person can change when feels being recorded wear Hence, we investigate different methodology assess physical status - integrating sensors tangible toy commonly used an...