• Publications
  • Influence
Enhancing the performance of adaptive iterative learning control with reinforcement learning
TLDR
A new adaptive ILC scheme is proposed, where the adaptation is supervised by reinforcement learning, and it is shown how to apply ILC to orientational motion, taking into account the curved geometry of SO(3). Expand
Subjectivity-based adjective ordering maximizes communicative success
TLDR
This work claims that pressures from successful reference resolution and the hierarchical structure of modification explain subjectivity-based ordering preferences and provides further support for this claim using large-scale simulations of reference scenarios, together with an empiricallymotivated adjective semantics. Expand
Learning of Exception Strategies in Assembly Tasks
TLDR
This work proposes a concurrent LfD framework, which associates demonstrated exception strategies to the given context and generates an appropriate policy that solves the assembly issue whenever a failure occurs and was validated in a peg-in-hole (PiH) task using Franka-Emika Panda robot. Expand
Learning of Robotic Throwing at a Target using a Qualitative Learning Reward
TLDR
Results show that learning with a simplified reward function that practically assigns a qualitative reward, just as a person would, can still be effectively used for RL using PoWER. Expand
Autonomous Learning of Assembly Tasks from the Corresponding Disassembly Tasks
TLDR
This work proposes to use hierarchical reinforcement learning, where learning is decomposed into a high-level decision-making and underlying lower-level intelligent compliant controller, which exploits the natural motion in a constrained environment. Expand
Incremental Policy Refinement by Recursive Regression and Kinesthetic Guidance
TLDR
This work proposes an incremental LfD framework that efficiently solves the issue of how to address the adaptation of existing policies and has been implemented and tested on a number of popular collaborative robots. Expand
Is it Living? Insights from Modeling Event-Oriented, Self-Motivated, Acting, Learning and Conversing Game Agents
TLDR
The work offers an approach to develop and thus to ground conceptual, semantic world knowledge in sensorimotor interactions and to couple this knowledge with a language to generate and comprehend language about the agent’s virtual world meaningfully. Expand
A Voice User Interface for Human-Robot Interaction on a Service Robot
Human-robot interaction is an important area of robotics. Traditional humanmachine and human-computer interfaces do not suffice for all use cases of mobile robots, and in particular humanoid robots.Expand
Modular Real-Time System for Upper-Body Motion Imitation on Humanoid Robot Talos
TLDR
Real-time motion transfer from a human demonstrator to an advanced humanoid robot Talos is presented and Talos can safely imitate upper-body human motion in real-time without significant delays. Expand
...
1
2
...