Embodied Artificial Intelligence Systems for UAVs

Dear Colleagues,

Unmanned aerial vehicles (UAVs) are on the verge of becoming an integral part of our lives, and their market is predicted to witness remarkable growth between 2023 and 2030. The continuous proliferation of UAVs and the evolving complexity of tasks impose significant demands on the computing and communication infrastructure, which are resource-constrained in terms of UAV computing power and UAV-to-cloud communication bandwidth. The miniaturization of UAV form factors, dynamic oper-ating scenarios, higher performance requirements, and cybersecurity considerations further exacerbate these demands. Edge computing is a computing paradigm where computation is performed mostly on power-constrained devices with limited pro-cessing capability or on-premise data centers. By pushing computing resources to the edge, closer to UAVs, edge computing en-ables low-latency service delivery for safety and mission-critical UAV applications.

This Special Issue aims to publish the latest contributions in developing Embodied Artificial Intelligence (EAI) software and hardware for mobile edge computing for UAVs to advance UAVs’ real-time, energy-efficient, adaptative, reliable, reconfigura-ble, and predictable performance. Researchers, developers, and industry practitioners working in this area are invited to present their views on the current trends, challenges, and state-of-the-art solutions addressing various challenges and issues in mobile edge computing for UAVs.

The recent development of EAI enables various types of robots to learn from their operating environments and self-improve (https://cacm.acm.org/blogcacm/building-computing-systems-for-embodied-artificial-intelligence/). Particularly for drones, the integration of EAI can empower them to perceive, process, and react to their environments as a living organism interacts with its surroundings. This application of AI makes drones more autonomous and efficient in various tasks. Some key applications in-clude:

1. Autonomous Navigation and Flight: Embodied AI allows drones to navigate complex environments without human inter-vention. Drones can detect and avoid obstacles by processing real-time sensory data, adjusting their flight paths, and main-taining stability in varying weather conditions.
2. Search and Rescue Operations: In search and rescue missions, drones equipped with embodied AI can scan large areas and identify signs of survivors using visual or heat signatures. These drones can operate in challenging terrains where human teams might not reach quickly or safely.
3. Agricultural Monitoring and Management: Drones with AI capabilities can monitor crop health, soil conditions, and mois-ture levels through sensors and provide precise data to farmers. This helps make informed decisions about irrigation, pes-ticide application, and harvesting, leading to increased crop yield and reduced resource waste.
4. Environmental Monitoring: AI-enabled drones are used for environmental conservation tasks, such as tracking wildlife, monitoring deforestation, and assessing pollution levels in air and water. They can autonomously collect data over large areas, providing valuable insights for environmental protection efforts.
5. Delivery and Logistics: AI-equipped drones can perform autonomous deliveries, navigating between complex urban land-scapes and optimizing delivery routes. This reduces delivery times and costs and increases efficiency in the logistics sector.

The development of these applications involves advancements in machine learning, computer vision, sensor technology, and robotics, making drones more adaptable and smarter in performing tasks that were traditionally challenging or hazardous for humans. These innovations continue to push the boundaries of what autonomous systems can achieve, promising significant impacts across multiple industries.

We welcome submissions from, but are not limited to, the following:

EAI Algorithms for Drones:
- Efficient perception, mapping, localization, motion planning, and control for EAI-enabled UAVs;
- Machine learning, deep learning, end-to-end learning, and federated learning for EAI-enabled UAVs;
- Application case studies for EAI-enabled UAVs.

EAI Systems for Drones:
- Resource management and computational offloading for mobile edge computing-enabled UAVs;
- Deployment frameworks and models for mobile edge computing-enabled UAVs.

EAI Computing Acceleration Hardware for Drones:
- Acceleration of UAV EAI applications with nontraditional computing hardware (e.g., ASICs, FPGAs, GPUs);
- Reconfigurable and self-adaptive computing design for open and evolving UAV environments;
- Performance and efficiency evaluation, analysis, and benchmarks for mobile edge computing-enabled UAVs.

Security and reliability:
- Security and privacy considerations for mobile edge computing-enabled UAV;
- Robustness and resilience considerations for mobile edge computing-enabled UAV.

Dr. Shaoshan Liu
Dr. Bo Yu
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Drones is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• edge computing
• energy efficiency and real-time performance
• hardware acceleration
• robustness and privacy
• resource management
• algorithm hardware codesign