A5046447340- Robotic Locomotion and Control
- Robot Manipulation and Learning
- Prosthetics and Rehabilitation Robotics
- Robotic Path Planning Algorithms
- Robotic Mechanisms and Dynamics
- Robotics and Automated Systems
- Modular Robots and Swarm Intelligence
- Real-time simulation and control systems
- Robotics and Sensor-Based Localization
- Soft Robotics and Applications
- Neurogenetic and Muscular Disorders Research
- Soil Mechanics and Vehicle Dynamics
- Manufacturing Process and Optimization
- Advanced Vision and Imaging
- Real-Time Systems Scheduling
- Space Satellite Systems and Control
- Context-Aware Activity Recognition Systems
- Embedded Systems Design Techniques
- Human Pose and Action Recognition
- Control and Dynamics of Mobile Robots
- Aerospace Engineering and Energy Systems
- Video Surveillance and Tracking Methods
- Teleoperation and Haptic Systems
- Winter Sports Injuries and Performance
- Software Testing and Debugging Techniques
Italian Institute of Technology
2015-2025
Université de Lorraine
2024-2025
Laboratoire Lorrain de Recherche en Informatique et ses Applications
2024-2025
Centre National de la Recherche Scientifique
2024-2025
Centre Inria de l'Université de Lorraine
2023-2024
Coral Reef Research Foundation
2022-2023
Nerviano Medical Sciences
2023
Centro Regional de Derechos Humanos y Justicia de Género, Corporación Humanas
2021
Institute of Electrical and Electronics Engineers
2019
Gorgias Press (United States)
2019
In this work, we present WALK‐MAN, a humanoid platform that has been developed to operate in realistic unstructured environment, and demonstrate new skills including powerful manipulation, robust balanced locomotion, high‐strength capabilities, physical sturdiness. To enable these WALK‐MAN design actuation are based on the most recent advancements of series elastic actuator drives with unique performance features differentiate robot from previous state‐of‐the‐art compliant actuated robots....
Despite the development of a large number mobile manipulation robots, very few platforms can demonstrate required strength and mechanical sturdiness to accommodate needs real-world applications with high payload moderate/harsh physical interaction demands, e.g., in disaster-response scenarios or heavy logistics/collaborative tasks. In this letter, we introduce design wheeled-legged platform capable executing demanding tasks, demonstrating significant resilience while possessing body size...
In this work we introduce XBotCore (Cross-Bot-Core), a light-weight, Real-Time (RT) software platform for EtherCAT-based robots. is open-source and designed to be both an RT robot control framework middleware. It satisfies hard requirements, while ensuring 1 kHz loop even in complex Multi-Degree-Of-Freedom systems. provides simple easy-to-use middleware Application Programming Interface (API), non-RT frameworks. This API completely flexible with respect the user wants utilize. Moreover it...
A fundamental aspect of controlling humanoid robots lies in the capability to exploit whole body perform tasks. This work introduces a novel control library called OpenSoT. OpenSoT is combined with joint impedance create framework that can effectively generate complex motion behaviors for humanoids according needs interaction level gives an easy way implement tasks, constraints, bounds and solvers by providing common interfaces. We present mathematical foundation validate it on compliant...
Transferring the motion from a human operator to humanoid robot is crucial step enable robots learn and replicate movements. The ability retarget in realtime whole-body motions that are challenging for balance critical teleoperation. In this work, we design retargeting framework allows of operator, acquired by wearable capture suit, while maintaining balance. We introduce some dynamic filter forbid dangerous can make fall. validate our approach through several experiments on iCub robot,...
Mobile manipulation robots have great potential for roles in support of rescuers on disaster-response missions. Robots can operate places too dangerous humans and therefore assist accomplishing hazardous tasks while their human operators work at a safe distance. We developed system that consists the highly flexible Centauro robot suitable control interfaces, including an immersive telepresence suit support-operator controls offering different levels autonomy.
In this paper we present OpenSoT, an open-source, recently developed software library, that can be used to solve robotics related control problems in a flexible and easy way. OpenSoT includes high-level interfaces state-of-the-art algorithms for kinematic/dynamic modelling, quadratic programming optimization, cost functions constraints specification. is implemented C++ permits rapid prototyping of controllers fixed or floating base, highly redundant robots such as (but not limited to)...
This work introduces a framework for the Cartesian control of multi-legged, highly redundant robots. The proposed allows untrained user to perform complex motion tasks with robotics platforms by leveraging simple, auto-generated ROS-based interface. Contrary other frameworks (e.g. ROS MoveIt!), we focus on execution trajectories that are specified online, rather than planned in advance, as it is case, instance, tele-operation and locomotion tasks. Moreover, address problem generating such...
Performing a demanding manipulation task with multi-legged loco-manipulation platform may require the exploitation of multiple external contacts different environment surfaces for counteracting forces. This is case pushing heavy object, where grip forces at ground not be adequate and establishing leg against wall turns out to an effective solution problem. In order produce such behaviour, this letter presents control architecture that able freely exploit complexity perform actions, e.g....
This paper deals with the self-collision avoidance problem for humanoid robots in an efficient way. Self-collision is introduced as a constraint each task hierarchical Inverse Kinematic (IK) problem. Since number of link pairs which needs to be updated and checked self-collision, every control loop, large, novel concept Self-Collision Avoidance Focus Interest (SCAFoI) proposed. SCAFoIs permits predict dynamically select necessary online improve computation efficiency. For several SCAFoIs,...
With the purpose of achieving a desired interaction performance for our compliant humanoid robot (COMAN), in this paper we propose semi-autonomous control framework and evaluate it experimentally valve turning setup. The structure consists various modules interfaces to identify valve, locate front perform manipulation. manipulation module implements four motion primitives (Reach, Grasp, Rotate Disengage) realizes corresponding impedance profile each phase accomplish task. In direction,...
In this work we introduced a prioritized Cartesian impedance control under the framework of Quadratic Programming (QP) optimization. particular, present formulation which is simpler than full inverse dynamics, avoids any matrix pseudo-inversion, kinematics computation and considers strict priorities among tasks. Our based on QP optimization permitting to take into account also explicit inequality constraints. We compare in simulation tracking results obtained with classical algebraic...
In this paper, we consider the problem of generating appropriate motions for a torque-controlled humanoid robot that is assigned multi-contact loco-manipulation task, i.e., task requires to move within environment by repeatedly establishing and breaking multiple, non-coplanar contacts. To end, present complete planning control framework multi-limbed robotic systems, such as humanoids. The layer works offline consists two sequential modules: first, stance planner computes sequence feasible...
The advent of humanoids has brought new challenges in the real-world application. As a part ongoing efforts to foster functionality robot accommodating real environment, this paper introduces recent progress on door opening task with our compliant humanoid, CoMan. We presents task-prioritized impedance control framework for an upper body system that includes dual-arm, waist, two soft hands, and 3D camera. Aimed create desired responses open door, novel stiffness modulation method is...
The widespread use of robotics in new application domains beyond industrial workplace settings necessitates systems that demonstrate functionalities far exceeding those classical robotic machines. These emerging applications involve complex tasks vary and must be carried out within a partially unknown environment, requiring autonomy adaptability further increase the intricacy system software architecture. Meeting demands consequent complexity their control infrastructures can quickly...
Designing robotic manipulators for generic tasks while meeting specific requirements is a complex, iterative process involving mechanical design, simulation, control, and testing.New computational design tools are needed to simplify speed up such processes.This work presents an original formulation of the problem, tailored help with strong reachability requirements.The primary challenges addressed in this twofold.First, necessity consider both continuous quantities discrete...
The ongoing trend from mass-produced to mass-customized products with batch sizes as small a single unit has highlighted the need for highly adaptable robotic systems lower downtime maintenance. To address these demands, this article proposes development of novel reconfigurable collaborative robot (cobot), which potential open up many new scenarios within rapidly emerging flexible manufacturing environments. As technological contribution, we present complete hard- and software architecture...
The present work addresses the generation of a walking gait with automatic footstep placement for quadrupedal robot, within Linear Model Predictive Control framework. Existing has shown how this is only possible non-convex programming framework, finding solution which well-known to be very hard. We propose way formulate joint optimization problem as an approximate QP linear constraints, whose global optimum can quickly found off-the-shelf solvers. More specifically, done by introducing...
In this work we present the concept of a pilot interface to control humanoid robot on an abstract level in unknown environments. The environment is perceived with stereo camera system and then simplified into set environmental primitives. Based these primitives proposes affordances pilot. Affordances are represented as certainty functions over space end-effector poses. operates by selecting among proposed related action primitives, i.e. Object-Action Complexes (OACs). Before initiating...
Generating complex movements in redundant robots like humanoids is usually done by means of multi-task controllers based on quadratic programming, where a multitude tasks organized according to strict or soft priorities. Time-consuming tuning and expertise are required choose suitable task priorities, optimize their gains. Here, we automatically learn the controller configuration (soft priorities Convergence Gains), looking for solutions that track variety desired trajectories efficiently...
A software and control architecture for a humanoid robot is complex large project, that involves team of developers/researchers to be coordinated requires many hard design choices. If such project has done in very limited time, i.e. less than one year, more constraints are added concepts as modular design, code reusability API definition need used much possible. In this work we describe the developed Walk-Man, participant at Darpa Robotics Challenge. The challenge required execute different...