Aerial Field Robotics

TL;DR

The chapter surveys open research problems that constrain resilient autonomy in aerial field robotics, focusing on MAVs operating in real-world environments with constrained computation. It identifies bottlenecks across embodiment, perception, planning, and learning, including endurance, safe physical interaction, and robustness to perceptual degradation. It argues for sparse or map-free representations to enable long-term navigation and for low-latency, sensor-to-action planning, citing learning-based approaches and drone-racing-inspired methods as promising. The authors advocate sustained cross-disciplinary collaboration and the development of a unified perception-action loop to render aerial robots field-ready for diverse industrial and natural environments.

Abstract

Aerial field robotics research represents the domain of study that aims to equip unmanned aerial vehicles - and as it pertains to this chapter, specifically Micro Aerial Vehicles (MAVs)- with the ability to operate in real-life environments that present challenges to safe navigation. We present the key elements of autonomy for MAVs that are resilient to collisions and sensing degradation, while operating under constrained computational resources. We overview aspects of the state of the art, outline bottlenecks to resilient navigation autonomy, and overview the field-readiness of MAVs. We conclude with notable contributions and discuss considerations for future research that are essential for resilience in aerial robotics.