Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Course Outline
Introduction to Robotic Manipulation and Deep Learning
- Overview of manipulation tasks and core system components.
- Comparison of traditional vs. learning-based approaches.
- Applications of deep learning in perception, planning, and control.
Perception for Manipulation
- Visual sensing and object detection strategies for grasping.
- 3D vision, depth sensing, and point cloud processing techniques.
- Training CNNs for object localization and segmentation.
Grasp Planning and Detection
- Foundations of classical grasp planning algorithms.
- Learning grasp poses through data and simulation.
- Implementation of grasp detection networks (e.g., GGCNN, Dex-Net).
Control and Motion Planning
- Inverse kinematics and trajectory generation principles.
- Learning-based motion planning and imitation learning methods.
- Utilising reinforcement learning for manipulation control policies.
Integration with ROS 2 and Simulation Environments
- Configuration of ROS 2 nodes for perception and control.
- Simulation of robotic manipulators in Gazebo and Isaac Sim.
- Integration of neural models for real-time control performance.
End-to-End Learning for Manipulation
- Unifying perception, policy, and control within comprehensive networks.
- Leveraging demonstration data for supervised policy learning.
- Domain adaptation strategies between simulation and real hardware.
Evaluation and Optimization
- Metrics for assessing grasp success, stability, and precision.
- Testing robustness under varying conditions and disturbances.
- Model compression and deployment strategies for edge devices.
Hands-on Project: Deep Learning-Based Robotic Grasping
- Designing a comprehensive perception-to-action pipeline.
- Training and testing a dedicated grasp detection model.
- Integrating the developed model into a simulated robotic arm.
Requirements
- A robust understanding of robotics kinematics and dynamics.
- Proficiency in Python and relevant deep learning frameworks.
- Working familiarity with ROS or equivalent robotic middleware.
Target Audience
- Robotics engineers designing intelligent manipulation systems.
- Perception and control specialists focused on grasping applications.
- Researchers and advanced practitioners specialising in robot learning and AI-based control.
28 Hours
Testimonials (2)
Supply of the materials (virtual machine) to get straight into the excersises, and the explanation of the Ros2 core. Why things work a certain way.
Arjan Bakema
Course - Autonomous Navigation & SLAM with ROS 2
its knowledge and utilization of AI for Robotics in the Future.
