DC2 - Methodology for Generic Evaluation of Intelligent Collaborative Robotic Systems

Objectives

1. To conduct research on dynamic evaluation methodologies, benchmarks, and protocols that address the adaptive, AI-driven capabilities of cognitive robots in collaborative environments, expanding beyond existing safety standards like ISO 10218 and TS 15066 that focus on rather static robot deployments.
2. o analyse how AI-driven perception and decision-making affect error propagation and control robustness in collaborative robots, and develop enhanced, adaptive safety mechanisms to ensure reliable and human-centric collaboration.

Expected Results

1. A comprehensive and validated testing framework that enables rigorous assessment of cognitive robots in collaborative environments. This framework should allow for the systematic evaluation of performance measures related to safety and error management. The results shall inform future safety regulation and testing protocols for applications of cognitive robots.
2. AI-driven control laws that demonstrate improved safety features and robust error management mechanisms. These control laws should reduce error propagation in real-time operation, leading to safer human-robot interactions and establishing best practices for incorporating AI in cobotics.

Planned Secondment(s)

S3: UBielefeld, Dr. Wrede, 2 months in M14-15
S4: Cellumation, Dr. Isken, 2 months in M26-27

Required Skills

Essential

• Degree in Computer Science or robotics
• Solid programming skills (Python), experience in Linux
• Good organisational skills.

Desirable

• Experience in system benchmark or testing
• Knowledge of ROS, simulation. Good understanding of Machine Learning
• Basic knowledge of embedded systems and electronics