Robust Control Technology in Low-Altitude Aerial Vehicle

课程简介

/ Indicative Syllabus Advanced Sensing Technologies for Low-Altitude Aerial Vehicles: Overview of Sensory Technology: Detailed study of various sensory technologies such as GNSS, LiDAR, visual sensors, mmWave radars, thermal cameras, and depth sensors, emphasizing their role in enhancing vehicle navigation and obstacle detection. Operational Integration: Techniques for the integration and calibration of these instruments to achieve optimal functionality in various environmental conditions, including urban and rural landscapes. Data Integration and Analysis Techniques for Aerial Systems: Data Handling Techniques: Exploration of methods for managing large volumes of data from multiple sensors, focusing on synchronization, -- 1 of 4 -- 2 alignment, and real-time processing challenges. Advanced Analysis: Use of sophisticated algorithms for noise reduction, feature extraction, and the combination of data sources to create comprehensive environmental models, enhancing decision-making processes. Computational Tools and Methodologies for Aerial Systems: Toolset Proficiency: In-depth training in the use of modern programming language for developing and testing machine learning models. Introduction to software environments and libraries specifically suited for aerial data analysis. Learning Techniques: Detailed examination of supervised, unsupervised, and reinforcement learning paradigms and their applicability to tasks such as predictive modeling, anomaly detection, and adaptive perception in low-altitude contexts. Robust Scene Perception in Dynamic Environments: Scene Analysis: Comprehensive coverage of algorithms for dynamic scene understanding, including visual odometry for tracking vehicle movement and 3D mapping to create detailed environmental representations. Algorithmic Challenges: Discussion of the challenges in implementing these algorithms in low-altitude scenarios, such as dealing with variable lighting conditions, weather impacts, and moving obstacles. Practical Applications in Low-Altitude Aerial Systems: Data-Driven Solutions: Focuses on applying techniques such as pattern discovery, topic modeling, and predictive analytics to enhance navigation and operational decisions in low-altitude environments. Practical Considerations: Discusses scalability and interpretability of machine learning models that are crucial for real-time applications in low-altitude systems. Also, addresses legal, social, and ethical considerations in deploying data analytics in this context.

目标

The objectives of this subject are to: 1. introduce the principles, concepts and models of most popular spatial perception algorithms for low-altitude aerial vehicles. 2. provide both theoretical and practical understanding of spatial perception models for low-altitude aerial vehicles such as machine learning and deep learning. 3. develop proficiency in designing, training, and optimizing spatial perception models for low-altitude aerial vehicles using modern programming language.