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
Foundations of Robotic Manipulation and Deep Learning
- A survey of manipulation tasks and core system components.
- Comparing traditional methods with learning-based approaches.
- The role of deep learning in perception, planning, and control loops.
Perception Capabilities for Manipulation
- Visual sensing techniques and object detection strategies for grasping.
- 3D vision, depth sensing technologies, and point cloud data processing.
- Training Convolutional Neural Networks (CNNs) for object localization and segmentation.
Grasp Planning and Detection Mechanisms
- Review of classical grasp planning algorithms.
- Learning grasp poses through data-driven and simulation-based methods.
- Implementation of grasp detection networks, including examples like GGCNN and Dex-Net.
Control Systems and Motion Planning
- Mastering inverse kinematics and trajectory generation.
- Applying learning-based motion planning and imitation learning techniques.
- Utilizing reinforcement learning for developing manipulation control policies.
Integration with ROS 2 and Simulation Platforms
- Configuration of ROS 2 nodes for perception and control tasks.
- Simulating robotic manipulators using Gazebo and Isaac Sim.
- Integrating neural models to achieve real-time control performance.
End-to-End Learning for Manipulation Tasks
- Unifying perception, policy generation, and control within integrated networks.
- Leveraging demonstration data for supervised policy learning.
- Addressing domain adaptation challenges between simulation and physical hardware.
Evaluation Metrics and System Optimization
- Defining metrics for grasp success, stability, and precision.
- Testing system robustness under varying conditions and disturbances.
- Model compression strategies and deployment on edge devices.
Practical Project: Deep Learning-Driven Robotic Grasping
- Architecting a complete perception-to-action pipeline.
- Training and validating a grasp detection model.
- Embedding the model into a simulated robotic arm for full-system testing.
Requirements
- A robust command of robotics kinematics and dynamics.
- Proficiency in Python and experience with major deep learning frameworks.
- Working knowledge of ROS or comparable robotic middleware systems.
Target Audience
- Robotics engineers focused on building intelligent manipulation systems.
- Specialists in perception and control disciplines engaged in grasping applications.
- Researchers and advanced practitioners specializing in robot learning and AI-driven 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.
