Aims and scope

The goal of Autonomous Robots is to report on the theory and applications of the computational aspects of robotic systems capable of some degree of self-sufficiency. Thus, the journal is aimed at the advances in robotics toward adaptation, autonomy, interaction, intelligence, manipulation, and mobility in an unstructured world. The term `robot' implies that the systems described here are capable of performing purposeful behaviors in the real world. They obtain inputs from the world through sensors and act upon the world through actuators. The connection between sensor data and actuation may result from advanced sensor data interpretation or it may involve complex decision making, goal interpretation and other aspects of reasoning. Most autonomous systems substantially change the environment they perceive, whether it be through navigation, manipulation or a combination of both.

Publication preference will be given to papers which include performance data on physical robots in the real world. In order to be considered for publication, papers which include only mathematical or simulation results must describe in sufficient detail a path for the transition from theory or simulation to the real world.

Papers published in these pages will report on original research in such areas as:

Computational architectures for autonomous systems
Human-robot interaction
Learning and adaptation in robots
Manipulation and locomotion
Multi-robot systems
Planning and navigation
Studies of autonomous robot systems
Sensing and perception
Self-calibration and self-repair for robots

Special Issue on Leveraging Implicit Representations for Learning-Enabled Autonomous Flight
Submission Date: 2026-03-01

This Topical Collection explores the integration of implicit representations within learning-enabled autonomous flight to improve adaptability, safety, and robustness in complex and dynamic real-world environments. Traditional modular pipelines relying on explicit models and manual specifications struggle with scalability and generalization, while purely data-driven end-to-end learning demands excessive amounts of data. Implicit methods offer a principled alternative by embedding physical knowledge, constraints, and goal specifications into machine learning frameworks across perception, planning, and control, bridging classical model-based and data-driven approaches. Demonstrating their effectiveness through hardware experiments and field deployments is essential for validating real-world applicability beyond simulations and controlled laboratory settings.

We invite submissions on implicit learning that can be utilized for aerial robots. Submissions should address theoretical concepts and practical challenges, especially emphasizing real-world applications and hardware deployments.

Topics to be covered include, but are not limited to:

    Implicit scene representations for aerial navigation
    Learning implicit dynamics for agile flight
    Implicit neural control and policy learning
    Implicit goal specification through large language models (LLMs) 1
    Human-robot interaction via implicit communication
    Optimization and planning with implicit networks
    Implicit models for supervised and self-supervised learning
    Reinforcement learning with implicit architectures
    Implicit behavioral cloning and imitation learning
    Benchmarking and comparative studies of implicit and explicit methods
    Learning implicit generative models for perception and control