Get in Touch
 Duration 21 hours

Course Outline

Introduction to Edge AI and TinyML

  • Overview of AI at the edge.
  • Advantages and challenges of executing AI on devices.
  • Applications in robotics and automation.

TinyML Fundamentals

  • Machine learning adapted for resource-constrained systems.
  • Model quantization, pruning, and compression techniques.
  • Supported frameworks and hardware platforms.

Model Development and Conversion

  • Training lightweight models using TensorFlow or PyTorch.
  • Converting models for TensorFlow Lite and PyTorch Mobile.
  • Testing and validating model accuracy.

Implementing On-Device Inference

  • Deploying AI models to embedded boards such as Arduino, Raspberry Pi, and Jetson Nano.
  • Integrating inference with robotic perception and control systems.
  • Executing real-time predictions and monitoring performance metrics.

Optimizing for Edge Performance

  • Minimizing latency and reducing energy consumption.
  • Utilizing hardware acceleration via NPUs and GPUs.
  • Benchmarking and profiling embedded inference.

Edge AI Frameworks and Tools

  • Working with TensorFlow Lite and Edge Impulse.
  • Exploring deployment options for PyTorch Mobile.
  • Debugging and tuning embedded ML workflows.

Practical Integration and Case Studies

  • Designing edge AI perception systems for robots.
  • Integrating TinyML with ROS-based robotics architectures.
  • Case studies covering autonomous navigation, object detection, and predictive maintenance.

Summary and Next Steps

Requirements

  • A solid understanding of embedded systems.
  • Proficiency in Python or C++ programming.
  • Familiarity with fundamental machine learning concepts.

Audience

  • Embedded developers.
  • Robotics engineers.
  • System integrators specializing in intelligent devices.

Number of participants


Price per participant (excl. VAT)

Testimonials (2)

Upcoming Courses

Related Categories