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 Edge AI in Industrial Environments
- The significance of edge computing in manufacturing processes
- Contrasting edge AI with cloud-based AI solutions
- Practical applications in computer vision, predictive maintenance, and control systems
Hardware Platforms and Device-Level Limitations
- Survey of popular edge hardware options (Raspberry Pi, NVIDIA Jetson, Intel NUC)
- Key factors regarding processing power, memory, and energy consumption
- Criteria for selecting the appropriate platform for specific applications
Model Development and Optimization for Edge Deployment
- Techniques for model compression, pruning, and quantization
- Utilizing TensorFlow Lite and ONNX for embedded systems
- Balancing model accuracy against speed in resource-constrained settings
Computer Vision and Sensor Fusion at the Edge
- Implementing edge-based visual inspection and monitoring workflows
- Aggregating data from diverse sensors (vibration, temperature, cameras)
- Performing real-time anomaly detection using Edge Impulse
Communication and Data Exchange Protocols
- Employing MQTT for industrial messaging standards
- Integration with SCADA, OPC-UA, and PLC systems
- Ensuring security and resilience in edge communications
Deployment and Field Testing Procedures
- Packaging and releasing models on edge devices
- Monitoring performance metrics and managing software updates
- Case study analysis: Real-time decision loops with local actuation
Scaling and Maintenance of Edge AI Systems
- Strategies for managing distributed edge devices
- Handling remote updates and model retraining cycles
- Considering lifecycle management for industrial-grade deployments
Summary and Recommended Next Steps
Requirements
- Foundational knowledge of embedded systems or IoT architectures
- Programming proficiency in Python or C/C++
- Experience with developing machine learning models
Target Audience
- Embedded software developers
- Industrial IoT teams
21 Hours
Testimonials (1)
That we can cover advance topic and work with real-life example
