This paper introduces SPIRAL (Self-Play Incremental Racing Algorithm for Learning), a novel approach for training autonomous drones in multi-agent racing competitions. SPIRAL distinctively employs a self-play mechanism to incrementally cultivate complex racing behaviors within a challenging, dynamic environment. Through this self-play core, drones continuously compete against increasingly proficient versions of themselves, naturally escalating the difficulty of competitive interactions. This progressive learning journey guides agents from mastering fundamental flight control to executing sophisticated cooperative multi-drone racing strategies. Our method is designed for versatility, allowing integration with any state-of-the-art Deep Reinforcement Learning (DRL) algorithms within its self-play framework. Simulations demonstrate the significant advantages of SPIRAL and benchmark the performance of various DRL algorithms operating within it. Consequently, we contribute a versatile, scalable, and self-improving learning framework to the field of autonomous drone racing. SPIRAL's capacity to autonomously generate appropriate and escalating challenges through its self-play dynamic offers a promising direction for developing robust and adaptive racing strategies in multi-agent environments. This research opens new avenues for enhancing the performance and reliability of autonomous racing drones in increasingly complex and competitive scenarios.

Autonomous drone racing has gained attention for its potential to push the boundaries of drone navigation technologies. While much of the existing research focuses on racing in obstacle-free environments, few studies have addressed the complexities of obstacle-aware racing, and approaches presented in these studies often suffer from overfitting, with learned policies generalizing poorly to new environments. This work addresses the challenge of developing a generalizable obstacle-aware drone racing policy using deep reinforcement learning. We propose applying domain randomization on racing tracks and obstacle configurations before every rollout, combined with parallel experience collection in randomized environments to achieve the goal. The proposed randomization strategy is shown to be effective through simulated experiments where drones reach speeds of up to 70 km/h, racing in unseen cluttered environments. This study serves as a stepping stone toward learning robust policies for obstacle-aware drone racing and general-purpose drone navigation in cluttered environments. Code is available at https://github.com/ErcBunny/IsaacGymEnvs.

In autonomous and mobile robotics, a principal challenge is resilient real-time environmental perception, particularly in situations characterized by unknown and dynamic elements, as exemplified in the context of autonomous drone racing. This study introduces a perception technique for detecting drone racing gates under illumination variations, which is common during high-speed drone flights. The proposed technique relies upon a lightweight neural network backbone augmented with capabilities for continual learning. The envisaged approach amalgamates predictions of the gates' positional coordinates, distance, and orientation, encapsulating them into a cohesive pose tuple. A comprehensive number of tests serve to underscore the efficacy of this approach in confronting diverse and challenging scenarios, specifically those involving variable lighting conditions. The proposed methodology exhibits notable robustness in the face of illumination variations, thereby substantiating its effectiveness.

It is often necessary for drones to complete delivery, photography, and rescue in the shortest time to increase efficiency. Many autonomous drone races provide platforms to pursue algorithms to finish races as quickly as possible for the above purpose. Unfortunately, existing methods often fail to keep training and racing time short in drone racing competitions. This motivates us to develop a high-efficient learning method by imitating the training experience of top racing drivers. Unlike traditional iterative learning control methods for accurate tracking, the proposed approach iteratively learns a trajectory online to finish the race as quickly as possible. Simulations and experiments using different models show that the proposed approach is model-free and is able to achieve the optimal result with low computation requirements. Furthermore, this approach surpasses some state-of-the-art methods in racing time on a benchmark drone racing platform. An experiment on a real quadcopter is also performed to demonstrate its effectiveness.

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.

Autonomous drones can operate in remote and unstructured environments, enabling various real-world applications. However, the lack of effective vision-based algorithms has been a stumbling block to achieving this goal. Existing systems often require hand-engineered components for state estimation, planning, and control. Such a sequential design involves laborious tuning, human heuristics, and compounding delays and errors. This paper tackles the vision-based autonomous-drone-racing problem by learning deep sensorimotor policies. We use contrastive learning to extract robust feature representations from the input images and leverage a two-stage learning-by-cheating framework for training a neural network policy. The resulting policy directly infers control commands with feature representations learned from raw images, forgoing the need for globally-consistent state estimation, trajectory planning, and handcrafted control design. Our experimental results indicate that our vision-based policy can achieve the same level of racing performance as the state-based policy while being robust against different visual disturbances and distractors. We believe this work serves as a stepping-stone toward developing intelligent vision-based autonomous systems that control the drone purely from image inputs, like human pilots.

In autonomous and mobile robotics, one of the main challenges is the robust on-the-fly perception of the environment, which is often unknown and dynamic, like in autonomous drone racing. In this work, we propose a novel deep neural network-based perception method for racing gate detection -- PencilNet -- which relies on a lightweight neural network backbone on top of a pencil filter. This approach unifies predictions of the gates' 2D position, distance, and orientation in a single pose tuple. We show that our method is effective for zero-shot sim-to-real transfer learning that does not need any real-world training samples. Moreover, our framework is highly robust to illumination changes commonly seen under rapid flight compared to state-of-art methods. A thorough set of experiments demonstrates the effectiveness of this approach in multiple challenging scenarios, where the drone completes various tracks under different lighting conditions.

Recently @Drone Racing League and @Lockheed Martin visited Austin, Texas as part of a series of drone races that pitted man against machine. The AIRR racing series stands for Artificial Intelligence Robotic Racing and took place this fall in four US cities. This drone racing series brought together teams of programmers from around the world to compete for a one million dollar prize. Each team was given an identical drone to work with and had to program it to complete a course using code only as its pilot - no human interaction at all with the drone. And the winner of each AI race then had to race against a human pilot, in this case @Gab707 from @Drone Racing League This entire event is part of the @Lockheed Martin AlphaPilot program, designed to foster innovation in the artificial intelligence and aviation worlds.