Aerobatic maneuvers in insect-scale flapping-wing aerial robots via deep-learned robust tube model predictive control

2025-08-04

AI Summary

Insect-scale flapping-wing robots face significant challenges in achieving insect-like, highly dynamic, and robust maneuverability due to their low inertia, aerodynamic uncertainties, and environmental sensitivity. To address this, we propose a deep learning–enhanced Robust Tube Model Predictive Control (RTMPC) framework that integrates imitation learning with a biologically inspired, two-layer fully connected neural network—mimicking central nervous system architecture—to enable 10-kHz real-time closed-loop feedback and force-mapping error compensation. Our approach achieves, for the first time on a sub-gram robot: (i) ten consecutive somersaults within 10 seconds; and (ii) agile yaw turns under wind disturbances up to 160 cm/s. The robot attains a lateral velocity of 197 cm/s and lateral acceleration of 11.7 m/s²—representing improvements of 447% and 255%, respectively, over prior state-of-the-art. This work establishes the most complex maneuver repertoire demonstrated to date for micro air vehicles.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Aerial insects exhibit highly agile maneuvers such as sharp braking, saccades, and body flips under disturbance. In contrast, insect-scale aerial robots are limited to tracking non-aggressive trajectories with small body acceleration. This performance gap is contributed by a combination of low robot inertia, fast dynamics, uncertainty in flapping-wing aerodynamics, and high susceptibility to environmental disturbance. Executing highly dynamic maneuvers requires the generation of aggressive flight trajectories that push against the hardware limit and a high-rate feedback controller that accounts for model and environmental uncertainty. Here, through designing a deep-learned robust tube model predictive controller, we showcase insect-like flight agility and robustness in a 750-millgram flapping-wing robot. Our model predictive controller can track aggressive flight trajectories under disturbance. To achieve a high feedback rate in a compute-constrained real-time system, we design imitation learning methods to train a two-layer, fully connected neural network, which resembles insect flight control architecture consisting of central nervous system and motor neurons. Our robot demonstrates insect-like saccade movements with lateral speed and acceleration of 197 centimeters per second and 11.7 meters per second square, representing 447$%$ and 255$%$ improvement over prior results. The robot can also perform saccade maneuvers under 160 centimeters per second wind disturbance and large command-to-force mapping errors. Furthermore, it performs 10 consecutive body flips in 11 seconds - the most challenging maneuver among sub-gram flyers. These results represent a milestone in achieving insect-scale flight agility and inspire future investigations on sensing and compute autonomy.

Problem

• Achieving insect-like agile maneuvers in flapping-wing robots
• Overcoming model and environmental uncertainty in flight control
• Enabling high-rate feedback control in compute-constrained systems

Innovation

• Deep-learned robust tube model predictive control
• Imitation learning for high-rate feedback
• Two-layer neural network mimicking insect flight