Litcius/Paper detail

A Sim-to-Real Deep Learning-Based Framework for Autonomous Nano-Drone Racing

Lorenzo Lamberti, Elia Cereda, Gabriele Abbate, Lorenzo Bellone, Victor Javier Kartsch Morinigo, Michał Barciś, Agata Barciś, Alessandro Giusti, Francesco Conti, Daniele Palossi

2024IEEE Robotics and Automation Letters13 citationsDOIOpen Access PDF

Abstract

Autonomous drone racing competitions are a proxy to improve unmanned aerial vehicles' perception, planning, and control skills. The recent emergence of autonomous nano-sized drone racing imposes new challenges, as their <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{10} \,\text{c}\text{m}$</tex-math></inline-formula> form factor heavily restricts the resources available onboard, including memory, computation, and sensors. This letter describes the methodology and technical implementation of the system winning the first autonomous nano-drone racing international competition: the “IMAV 2022 Nanocopter AI Challenge.” We developed a fully onboard deep learning approach for visual navigation trained only on simulation images to achieve this goal. Our approach includes a convolutional neural network for obstacle avoidance, a sim-to-real dataset collection procedure, and a navigation policy that we selected, characterized, and adapted through simulation and actual in-field experiments. Our system ranked <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1\text{st}$</tex-math></inline-formula> among six competing teams at the competition. In our best attempt, we scored <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{115} \,\text{m}$</tex-math></inline-formula> of traveled distance in the allotted 5-minute flight, never crashing while dodging static and dynamic obstacles. Sharing our knowledge with the research community, we aim to provide a solid groundwork to foster future development in this field.

Topics & Concepts

DroneObstacle avoidanceComputer scienceField (mathematics)Convolutional neural networkArtificial intelligenceObstacleDeep learningCompetition (biology)Proxy (statistics)AeronauticsSimulationMachine learningRobotEngineeringMobile robotGeographyPure mathematicsBiologyGeneticsMathematicsEcologyArchaeologyRobotic Path Planning AlgorithmsRobotics and Sensor-Based LocalizationUAV Applications and Optimization