Autonomous Drone Racing: A Survey
arXiv:2301.01755 · doi:10.1109/TRO.2024.3400838
Abstract
Over the last decade, the use of autonomous drone systems for surveying, search and rescue, or last-mile delivery has increased exponentially. With the rise of these applications comes the need for highly robust, safety-critical algorithms which can operate drones in complex and uncertain environments. Additionally, flying fast enables drones to cover more ground which in turn increases productivity and further strengthens their use case. One proxy for developing algorithms used in high-speed navigation is the task of autonomous drone racing, where researchers program drones to fly through a sequence of gates and avoid obstacles as quickly as possible using onboard sensors and limited computational power. Speeds and accelerations exceed over 80 kph and 4 g respectively, raising significant challenges across perception, planning, control, and state estimation. To achieve maximum performance, systems require real-time algorithms that are robust to motion blur, high dynamic range, model uncertainties, aerodynamic disturbances, and often unpredictable opponents. This survey covers the progression of autonomous drone racing across model-based and learning-based approaches. We provide an overview of the field, its evolution over the years, and conclude with the biggest challenges and open questions to be faced in the future.
26 pages
References in corpus (50)
- VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator
- Event-based Vision: A Survey
- DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
- Optimization-Based Autonomous Racing of 1:43 Scale RC Cars
- Ultimate SLAM? Combining Events, Images, and IMU for Robust Visual SLAM in HDR and High Speed Scenarios
- Differential Flatness of Quadrotor Dynamics Subject to Rotor Drag for Accurate Tracking of High-Speed Trajectories
- Learning High-Speed Flight in the Wild
- Autonomous Vehicles on the Edge: A Survey on Autonomous Vehicle Racing
- Neural Lander: Stable Drone Landing Control using Learned Dynamics
- Deep Drone Racing: From Simulation to Reality with Domain Randomization
- Time-Optimal Planning for Quadrotor Waypoint Flight
- DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras
- Reaching the Limit in Autonomous Racing: Optimal Control versus Reinforcement Learning
- Fast, Autonomous Flight in GPS-Denied and Cluttered Environments
- Differentiable MPC for End-to-end Planning and Control
- Aggressive Quadrotor Flight through Narrow Gaps with Onboard Sensing and Computing using Active Vision
- Sim4CV: A Photo-Realistic Simulator for Computer Vision Applications
- The MRS UAV System: Pushing the Frontiers of Reproducible Research, Real-world Deployment, and Education with Autonomous Unmanned Aerial Vehicles
- FlightGoggles: A Modular Framework for Photorealistic Camera, Exteroceptive Sensor, and Dynamics Simulation
- Agilicious: Open-Source and Open-Hardware Agile Quadrotor for Vision-Based Flight
- NeuroBEM: Hybrid Aerodynamic Quadrotor Model
- Event-aided Direct Sparse Odometry
- ESVIO: Event-based Stereo Visual Inertial Odometry
- Autonomous Quadrotor Flight despite Rotor Failure with Onboard Vision Sensors: Frames vs. Events
- Autonomous drone race: A computationally efficient vision-based navigation and control strategy
- Nonlinear Model Predictive Control with Enhanced Actuator Model for Multi-Rotor Aerial Vehicles with Generic Designs
- PANTHER: Perception-Aware Trajectory Planner in Dynamic Environments
- Human-Piloted Drone Racing: Visual Processing and Control
- Learning Minimum-Time Flight in Cluttered Environments
- Visual-Inertial Odometry of Aerial Robots
- Theseus: A Library for Differentiable Nonlinear Optimization
- Visual Model-predictive Localization for Computationally Efficient Autonomous Racing of a 72-gram Drone
- Safe Reinforcement Learning via Curriculum Induction
- Deep Patch Visual Odometry
- Neural Moving Horizon Estimation for Robust Flight Control
- Perception-aware Path Planning
- PyPose: A Library for Robot Learning with Physics-based Optimization
- Visual Attention Prediction Improves Performance of Autonomous Drone Racing Agents
- Safe Reinforcement Learning Using Black-Box Reachability Analysis
- End-to-end Neural Network Based Quadcopter control
- Aggressive Online Control of a Quadrotor via Deep Network Representations of Optimality Principles
- The Artificial Intelligence behind the winning entry to the 2019 AI Robotic Racing Competition
- Towards Time-Optimal Tunnel-Following for Quadrotors
- A Sim-to-Real Deep Learning-based Framework for Autonomous Nano-drone Racing
- Race Against the Machine: a Fully-annotated, Open-design Dataset of Autonomous and Piloted High-speed Flight
- Aerial Gym -- Isaac Gym Simulator for Aerial Robots
- CNN-based Ego-Motion Estimation for Fast MAV Maneuvers
- Euclidean and non-Euclidean Trajectory Optimization Approaches for Quadrotor Racing
- MAVRL: Learn to Fly in Cluttered Environments with Varying Speed
- CUAHN-VIO: Content-and-Uncertainty-Aware Homography Network for Visual-Inertial Odometry
Cited by in corpus (14)
- A Sim-to-Real Deep Learning-based Framework for Autonomous Nano-drone Racing
- Race Against the Machine: a Fully-annotated, Open-design Dataset of Autonomous and Piloted High-speed Flight
- A Generalized Thrust Estimation and Control Approach for Multirotors Micro Aerial Vehicles
- Generalized Multi-Speed Dubins Motion Model
- The Mini Wheelbot: A Testbed for Learning-based Balancing, Flips, and Articulated Driving
- Tangled Program Graphs as an alternative to DRL-based control algorithms for UAVs
- LoL-NMPC: Low-Level Dynamics Integration in Nonlinear Model Predictive Control for Unmanned Aerial Vehicles
- On Your Own: Pro-level Autonomous Drone Racing in Uninstrumented Arenas
- Autonomous UAV Flight Navigation in Confined Spaces: A Reinforcement Learning Approach
- InCrowd-VI: A Realistic Visual-Inertial Dataset for Evaluating SLAM in Indoor Pedestrian-Rich Spaces for Human Navigation
- GRaD-Nav++: Vision-Language Model Enabled Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics
- Reinforcement Learning with Lie Group Orientations for Robotics
- Enhancing Drone Light Shows Performances: Optimal Allocation and Trajectories for Swarm Drone Formations
- Vision-Guided Outdoor Flight and Obstacle Evasion via Reinforcement Learning