Skip to main content

Department of Computer Science

Mitrevski

Dr Alex Mitrevski

Lecturer/ Institute for AI and Autonomous Systems (A2S)

Unit

Department of Computer Science, Institute for Artificial Intelligence and Autonomous Systems (A2S)

Location

Sankt Augustin

Room

C203

Address

Grantham-Allee 20

53757, Sankt Augustin

Telephone

+492241 865206

Profile

Research areas

  • Knowledge representation and reasoning (knowledge retrieval, forgetting mechanisms, template- and case-based reasoning)
  • Lifelong robot learning
  • Simulation-based robot learning and reasoning
  • Robot fault detection and diagnosis
  • Cognitive robotics

Teaching

Lecturer

  • Cognitive Robotics (SS24, SS23)
  • Robot Learning (WS24, WS23)
  • Autonomous Mobile Robots (WS24, SS24, WS23)
  • Software Engineering for Robotics (WS24, SS24, WS23)
  • Research and Development Colloquium (WS19; joint lecturer in SS18)
  • Fault Detection and Diagnosis (SS19)

Teaching assistant

  • Mathematics for Robotics and Control (SS21, WS20, SS20, WS19, WS18, SS18, WS17, joint TA in SS17)
  • Software Development Project  - project coach (WS22, SS22, SS21, WS20, SS20)
  • Scientific Experimentation and Evaluation (WS18, SS18, WS17, joint TA in SS17)

Supervised master's theses

  • On using conversational AI for autonomous planning goal specification
  • Visuomotor policy learning for predictive manipulation
  • Coupled active perception for mobile manipulation in unknown environments
  • Multimodal deep learning-based adaptation of robot behaviour for assistive robotics
  • Experience-based path planning framework for real-time learning from demonstration
  • Outdoor navigation in uneven terrain using learning methods
  • Robust robot task execution using monitoring and recovery in anomalous conditions
  • Multimodal deep learning-based adaptation of robot behaviour for assistive robotics
  • Robust environment sound classification and anomaly detection using deep learning
  • Towards improvements on RoboCup@Home robots architecture, capabilities and development process
  • Lifelong action learning for socially assistive robots

Supervised R&D projects

  • Incorporating contextual knowledge into human-robot collaborative task execution
  • Learning corrective models for multistep actions by analysing videos
  • Integrating anomaly detection with fault diagnosis for long-horizon tasks
  • Registering and visualizing point cloud data with existing 3D CityGML Models
  • A comparative analysis of fault detection approaches in mobile robots
  • Benchmarking object placement algorithms for mobile robotic manipulators
  • Tell your robot what to do: Evaluation of natural language models for robot command processing
  • Safe and fault-aware child-robot natural language interaction
  • Manipulating handles in domestic environments
  • Learning grasp evaluation models using synthetic 3D object-grasp representations
  • Dynamic motion primitives
  • Ontology-based robot fault diagnosis
  • Learning human gestures by imitation for robots
  • Benchmarking object placement algorithms for mobile robotic manipulators
  • Learning-based indoor robot navigation
  • Development and implementation of a self-learning control approach for contact-rich object manipulation
  • Automated test generation for robot self-examination
  • Explainability analysis for skill execution
  • Learning efficient knife handling skills in semi-structured environments on a dual-arm robot
  • Personalised behaviour models for child-robot interaction
  • Semantic information by acoustic clues: A modern approach to anomaly detection for robotics

Research Projects

KEROL

KEROL aims to develop (i) a data collection and annotation interface that makes it possible to execute various robot skills and annotate execution failures in those, and (ii) a simulation-based environment that can be used to evaluate the execution of learned skills under a variety of conditions.

Project management at the H-BRS

Dr Alex Mitrevski
MigrAVE

In the MigrAVE project, the objective is to develop supporting technologies for children diagnosed with Autism Spectrum Disorder (ASD). ASD is a disorder that leads to behavioural and social challenges for those affected, for instance difficulty recognising emotions, general inability to act appropriately in everyday social situations, or repetitive behaviours. Early treatment of ASD could mitigate the effects of the disorder and improve the quality of life of affected people.

Project management at the H-BRS

Professor im Ruhestand / retired professor Dr. Paul G. Plöger
ROPOD

Objectives Develop and implement a disruptive concept for automatically guided vehicles (AGVs) that lowers the still existing barrier in logistics by offering • cost-effective, automated or semi-automated indoor transportation of goods, • while coping with existing legacy in terms of size, shape, and weight of goods and containers, • without imposing disruptive changes in existing logistic solutions, such as rebuilding entire warehouses or switching to new containers or storage technology.

Project management at the H-BRS

Prof. Dr Erwin Prassler

Publications