David Cuartielles, Malmo University, Sweden

Adel Mellit, University of Jijel, Algeria

Marcelo Rovai, Federal University of Itajuba, Brazil

Rosdiadee Nordin, Sunway University, Malaysia

Marco Zennaro, ICTP, Italy

The rapid evolution of Artificial Intelligence has traditionally relied on massive, centralized data centers. However, for many regions persistent cloud connectivity remains a luxury. This Special Collection on Edge AI for Development focuses on the paradigm shift toward “localized intelligence”—deploying AI algorithms directly on IoT devices, smartphones, and low-power hardware at the edge of the network. By moving computation closer to the data source, we can address critical barriers such as high latency, prohibitive bandwidth costs, and data privacy concerns. This collection seeks to highlight how Edge AI can empower marginalized communities, providing real-time insights in environments where traditional digital infrastructure is sparse or non-existent. We aim to bridge the gap between high-level algorithmic research and practical, field-ready deployments that drive social and economic impact.

We invite original research, case studies, and review articles that explore the intersection of decentralized AI and sustainable growth. Key Topics of Interest include, but are not limited to: Resource-Constrained Machine Learning: Techniques such as model pruning, quantization, and knowledge distillation for deployment on low-power microcontrollers. Precision Agriculture: Edge-based crop disease detection, automated irrigation systems, and livestock monitoring in remote areas. Global Health & Diagnostics: Real-time epidemic tracking, and wearable health monitors that function offline. Environmental Monitoring: Decentralized systems for illegal logging detection, wildlife tracking, and localized air/water quality sensing. Infrastructure & Energy: Smart grid management at the edge and decentralized waste management systems. Ethics and Governance: Privacy-preserving on-device learning (e.g., Federated Learning) and addressing algorithmic bias in localized datasets. Connectivity-Agnostic Solutions: Hybrid edge-cloud architectures designed specifically for intermittent or low-bandwidth environments. Solar power systems: Edge AI applications for solar power systems, including real-time fault detection and diagnosis, predictive maintenance, smart monitoring using TinyML models, smartphone inspection and image analysis.

The call for abstracts will be open from April 1st to July 31st 2026. Abstracts (300–500 words) should clearly outline the research focus and relevance to the journal’s scope/pillar, methodology, and interdisciplinary contribution, as well as anticipated findings or implications. Review of abstracts: A rolling review process will begin during the final weeks of the call. This will allow early submissions to receive feedback promptly. All reviews will be completed within one month of the call’s closing. Selected authors will receive invitations to submit full manuscripts. Manuscript development window: Authors will have until September 30th to submit manuscripts between 4,000 and 5,000 words. Manuscripts will then undergo peer review. Once authors receive their peer reviews, they will have a maximum of 2 weeks to send their final version. The special issue will be published in December 2026.

Abstract submission deadline: July 31, 2026
Full paper submission deadline: October 30, 2026
Publication date: December 2026