The Development of an Autonomous Multirotor Drone in Conjunction with OptiTrack; The Future of Autonomous Drones in Cities

Author:

orcid.org/0009-0004-9460-7387
McNabb, Luke, School of Engineering and Applied Science, University of Virginia

Advisors:
Furukawa, Tomonari, School of Mechanical and Aerospace Engineering, University of Virginia
Francisco, Pedro, Engineering and Society, School of Engineering and Applied Science, University of Virginia

Abstract:

The integration of drones into smart cities will create new urban development
opportunities in many areas including deliveries and public safety, but raises a new set of critical
questions regarding liability, regulation, and societal impact. In Dr. Wang's “Challenges and
Opportunities in Lidar Remote Sensing,” he highlights the many current shortcomings and
opportunities with current lidar (2021). First, normal weather events such as rain, fog, and dust
can interfere with the lidar’s laser prematurely before the laser could reach a real surface,
corrupting the scan. My technical research aimed to improve key sensors and algorithms to
improve autonomous drone flight. My STS research drew parallel from uavs to prior liability
laws and lawsuits of self-driving cars. The responsibility in drone accidents would shift towards
manufacturers, incentivizing them to prioritize safety measures.
The journey towards fully autonomous drones is marked by significant advancements in
navigation techniques and obstacle avoidance strategies. Optical flow and lidar represent two
primary schools of thought, each with its strengths and limitations. While optical flow excels in
relative motion perception and precise maneuvers, lidar offers a comprehensive 3D mapping
capability for enhanced environment perception.
My technical report provides a comprehensive overview of the design considerations for
creating an autonomous drone. A manually operated drone was constructed last semester, and
autonomous functions with mission planning were achieved this semester–first through
ArduPilot and in the future, through Raspberry Pi. The incorporation of advanced sensors,
computing power, and autonomy features will result in a versatile platform capable of
performing a wide range of tasks independently. This semester, we successfully established a
secure testing protocol within the reactor room by employing a pulley in order to effectively
minimize risks and enhance safety during manual testing procedures. After a few successful
outdoor GPS autonomous flights, the next step for this project is to establish a closed loop
system connecting the OptiTrack setup within the reactor room with the Raspberry Pi and
LiDAR sensors integrated into the drone.
The integration of drones into smart cities raises significant questions regarding liability
and regulatory frameworks. My STS research explores the evolving dynamics of responsibility
in the context of autonomous drones, drawing parallels with the emergence of self-driving cars to
shed light on potential legal and ethical challenges. Through an interdisciplinary approach
encompassing law, technology, and sociology, this research investigates the implications of
assigning liability to manufacturers versus individuals in scenarios of drone accidents and
disruptions. The analysis suggests that while a manufacturer’s liability will incentivize
companies to prioritize product reliability and safety, certain circumstances such as hacking or
sabotage would shift blame again.
In conclusion, as my technical report aimed to do, navigational technologies and
algorithms will improve, making drones more prevalent in many cities and industries like
shipping and safety. Both companies and individuals will require laws and regulations to govern
the use of autonomous drones.