香港現場AI for Robotics培訓

Edge AI enables artificial intelligence models to run directly on embedded or resource-constrained devices, reducing latency and power consumption while increasing autonomy and privacy in robotic systems.

This instructor-led, live training (online or onsite) is aimed at intermediate-level embedded developers and robotics engineers who wish to implement machine learning inference and optimization techniques directly on robotic hardware using TinyML and edge AI frameworks.

By the end of this training, participants will be able to:

• Understand the fundamentals of TinyML and edge AI for robotics.
• Convert and deploy AI models for on-device inference.
• Optimize models for speed, size, and energy efficiency.
• Integrate edge AI systems into robotic control architectures.
• Evaluate performance and accuracy in real-world scenarios.

Format of the Course

• Interactive lecture and discussion.
• Hands-on practice using TinyML and edge AI toolchains.
• Practical exercises on embedded and robotic hardware platforms.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

Reinforcement learning (RL) is a machine learning paradigm where agents learn to make decisions by interacting with an environment. In robotics, RL enables autonomous systems to develop adaptive control and decision-making capabilities through experience and feedback.

This instructor-led, live training (online or onsite) is aimed at advanced-level machine learning engineers, robotics researchers, and developers who wish to design, implement, and deploy reinforcement learning algorithms in robotic applications.

By the end of this training, participants will be able to:

• Understand the principles and mathematics of reinforcement learning.
• Implement RL algorithms such as Q-learning, DDPG, and PPO.
• Integrate RL with robotic simulation environments using OpenAI Gym and ROS 2.
• Train robots to perform complex tasks autonomously through trial and error.
• Optimize training performance using deep learning frameworks like PyTorch.

Format of the Course

• Interactive lecture and discussion.
• Hands-on implementation using Python, PyTorch, and OpenAI Gym.
• Practical exercises in simulated or physical robotic environments.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

OpenCV is an open-source computer vision library that enables real-time image processing, while deep learning frameworks such as TensorFlow provide the tools for intelligent perception and decision-making in robotic systems.

This instructor-led, live training (online or onsite) is aimed at intermediate-level robotics engineers, computer vision practitioners, and machine learning engineers who wish to apply computer vision and deep learning techniques for robotic perception and autonomy.

By the end of this training, participants will be able to:

• Implement computer vision pipelines using OpenCV.
• Integrate deep learning models for object detection and recognition.
• Use vision-based data for robotic control and navigation.
• Combine classical vision algorithms with deep neural networks.
• Deploy computer vision systems on embedded and robotic platforms.

Format of the Course

• Interactive lecture and discussion.
• Hands-on practice using OpenCV and TensorFlow.
• Live-lab implementation on simulated or physical robotic systems.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

ROS 2 (Robot Operating System 2) is an open-source framework designed to support the development of complex and scalable robotic applications.

This instructor-led, live training (online or onsite) is aimed at intermediate-level robotics engineers and developers who wish to implement autonomous navigation and SLAM (Simultaneous Localization and Mapping) using ROS 2.

By the end of this training, participants will be able to:

• Set up and configure ROS 2 for autonomous navigation applications.
• Implement SLAM algorithms for mapping and localization.
• Integrate sensors such as LiDAR and cameras with ROS 2.
• Simulate and test autonomous navigation in Gazebo.
• Deploy navigation stacks on physical robots.

Format of the Course

• Interactive lecture and discussion.
• Hands-on practice using ROS 2 tools and simulation environments.
• Live-lab implementation and testing on virtual or physical robots.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

Robotics 是人工智能（AI）中的一個領域，涉及智能和高效機器的編程與設計。

本課程爲講師指導的培訓（線上或線下），面向希望通過基本AI方法編程和創建機器人的工程師。

在本培訓結束時，參與者將能夠：

• 實現濾波器（卡爾曼和粒子濾波器），使機器人能夠定位環境中的移動物體。
• 實現搜索算法和運動規劃。
• 實現PID控制，以調節機器人在環境中的運動。
• 實現SLAM算法，使機器人能夠繪製未知環境的地圖。

課程形式

• 互動講座與討論。
• 大量練習與實踐。
• 在即時實驗室環境中進行實際操作。

課程定製選項

• 如需爲此課程定製培訓，請聯繫我們進行安排。

在這項由講師指導的香港（線上或線下）培訓中，學員將學習用於編程不同類型機器人的各種技術、框架和方法，這些機器人將應用於核技術和環境系統領域。

本課程爲期6周，每週5天，每天4小時。課程內容包括講座、討論以及在實驗室環境中的實際操作機器人開發。學員將完成各種與其工作相關的實際項目，以實踐所學知識。

本課程的目標硬件將通過模擬軟件進行3D模擬。編程機器人將使用ROS（機器人操作系統）開源框架、C++和Python。

培訓結束後，學員將能夠：

• 理解機器人技術中的關鍵概念。
• 理解並管理機器人系統中軟件與硬件的交互。
• 理解並實現支撐機器人技術的軟件組件。
• 構建並操作一個模擬的機械機器人，使其能夠通過視覺、感知、處理、導航和語音與人互動。
• 理解構建智能機器人所需的人工智能要素（機器學習、深度學習等）。
• 實現過濾器（卡爾曼和粒子）以使機器人能夠定位其環境中的移動物體。
• 實現搜索算法和運動規劃。
• 實現PID控制以調節機器人在環境中的運動。
• 實現SLAM算法以使機器人能夠繪製未知環境的地圖。
• 通過深度學習擴展機器人執行復雜任務的能力。
• 在現實場景中測試和排除機器人故障。

在這次由講師指導的線下或線上培訓中，學員將學習編程不同類型機器人所需的各種技術、框架和技巧，這些機器人將應用於核技術和環境系統領域。

本課程爲期4周，每週5天。每天4小時，包括講座、討論和在實驗室環境中的實際操作機器人開發。學員將完成各種適用於其工作的實際項目，以實踐所學的知識。

本課程的目標硬件將通過模擬軟件進行3D模擬。代碼隨後將加載到物理硬件（如Arduino或其他）上進行最終部署測試。將使用ROS（機器人操作系統）開源框架、C++和Python來編程機器人。

培訓結束後，學員將能夠：

• 理解機器人技術中的關鍵概念。
• 理解並管理機器人系統中軟件與硬件的交互。
• 理解並實現支撐機器人技術的軟件組件。
• 構建並操作一個模擬的機械機器人，使其能夠通過視覺、感知、處理、導航和語音與人類互動。
• 理解構建智能機器人所需的人工智能元素（如機器學習、深度學習等）。
• 實現濾波器（卡爾曼和粒子濾波器），使機器人能夠定位環境中的移動物體。
• 實現搜索算法和運動規劃。
• 實現PID控制，以調節機器人在環境中的運動。
• 實現SLAM算法，使機器人能夠繪製未知環境的地圖。
• 在真實場景中測試和調試機器人。