Perception and AI for Field Robotics

Dear Colleagues,

In recent decades, rapid advances in sensing technologies and artificial intelligence (AI) have significantly expanded robots’ capabilities to perceive, reason, and operate autonomously or semi-autonomously in complex, dynamic, and unstructured outdoor environments. The deployment of robots in such real-world scenarios—agricultural fields, mines, forests, construction sites, oceans, and critical infrastructure—is commonly referred to as field robotics.

Operating safely and efficiently in harsh, cluttered, and often unpredictable conditions requires robust perception pipelines built on advanced sensing systems (e.g., LiDAR, cameras, radar, GNSS/INS, and hyperspectral sensors). These systems must support reliable object detection, tracking, localization, mapping, and high-level scene understanding under varying illumination, weather, and terrain. In parallel, modern AI methods—ranging from deep learning to probabilistic reasoning and planning—have become central to many field robotics tasks, enabling robots to make informed decisions, handle uncertainty, and adapt to environmental changes over long time scales.

Despite substantial progress, perception and AI for field robotics still face critical challenges. These include dealing with sparse and noisy sensor data, extreme domain shifts between lab and field conditions, long-term autonomy, reliable multi-sensor fusion, limited onboard computation, safety and interpretability requirements, and the scarcity of standardized datasets and benchmarks for outdoor environments.

This Special Issue, titled “Perception and AI for Field Robotics”, aims to bring together scientists, researchers, and practitioners from academia, industry, and government to present original contributions that advance the state of the art. This issue will serve as a platform to showcase cutting-edge research, recent developments, emerging trends, and open challenges in perception and AI for field robotics.

Topics of interest include, but are not limited to, the following:

• Field robotics applications in agriculture, environmental monitoring, exploration, infrastructure inspection, public services, construction, mining, energy, and logistics;
• Multi-sensor fusion (e.g., LiDAR–camera–radar–GNSS/INS and multispectral/hyperspectral fusion) for robust perception;
• Sensor calibration, synchronization, and self-calibration in the field;
• In-field object detection, recognition, and classification;
• In-field object tracking and pose estimation under occlusions and changing environments;
• Localization and mapping (SLAM, semantic mapping, and multi-robot mapping) in outdoor and GPS-challenged environments;
• AI-based decision-making and planning for autonomous and semi-autonomous field robots;
• Learning under domain shift and adverse conditions (weather, lighting, and seasonal changes);
• Uncertainty modeling, safety, and interpretability in perception and decision-making pipelines;
• Resource-aware and real-time perception on embedded and edge computing platforms;
• Simulation, digital twins, and virtual testing for field robotics perception and AI;
• Datasets, benchmarks, and evaluation protocols for perception and AI in field robotics.

Dr. Cédric Pradalier
Dr. Tito Arevalo
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 250 words) can be sent to the Editorial Office for assessment.

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. Robotics 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 1800 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.

Publisher’s Notice

The Special Issue has been shifted from Section AI in Robotics to Section Agricultural and Field Robotics on 25 December 2025. At the time of the move, there were no publications in this Special Issue.

• field robotics
• multi-sensor fusion
• object detection
• localization and mapping
• AI-based decision-making and planning