Learning high-speed flight in the wild

Quadrotors are agile. Unlike most other machines, they can traverse extremely complex environments at high speeds. To date, only expert human pilots have been able to fully exploit their capabilities. Autonomous opera-tion with onboard sensing and computation has been limited to low speeds. State-of-the-art methods generally separate the navigation problem into subtasks: sensing, mapping, and planning. Although this approach has proven successful at low speeds, the separation it builds upon can be problematic for high-speed navigation in cluttered environments. The subtasks are executed sequentially, leading to increased processing latency and a compounding of errors through the pipeline. Here, we propose an end-to-end approach that can autonomously fly quadrotors through complex natural and human-made environments at high speeds with purely onboard sensing and computation. The key principle is to directly map noisy sensory observations to collision-free trajec-tories in a receding-horizon fashion. This direct mapping drastically reduces processing latency and increases ro-bustness to noisy and incomplete perception. The sensorimotor mapping is performed by a convolutional network that is trained exclusively in simulation via privileged learning: imitating an expert with access to privi-leged information. By simulating realistic sensor noise, our approach achieves zero-shot transfer from simulation to challenging real-world environments that were never experienced during training: dense forests, snow-covered terrain, derailed trains, and collapsed buildings. Our work demonstrates that end-to-end policies trained in simu-lation enable high-speed autonomous flight through challenging environments, outperforming traditional ob-stacle avoidance pipelines.