Description

This presentation introduces the fundamental ideas behind Reinforcement Learning (RL) in a simple and intuitive way. It first places RL within the broader field of Machine Learning, then explains the core agent–environment interaction loop, including concepts such as states, actions, rewards, policies, and learning through trial and error.

The presentation also includes several illustrative diagrams and conceptual examples to make the main ideas easier to understand. After the theoretical introduction, three practical case studies are presented: leader–follower navigation with collision avoidance, block pushing, and autonomous drifting. As an additional extension, the presentation briefly discusses the use of a UR5 robotic manipulator combined with Vision-Language Models for perception-guided robotic tasks.

Schedule

• Timezone: Europe/Berlin
• Start date: 2026-05-14
• End date: 2026-05-19
• Repeat frequency: Once
• Time slots: 11:30 – 12:10

Core details

Contact