Evolution of Robust High Speed Optical-Flow-Based Landing for Autonomous MAVs

TL;DR

This paper presents an evolutionary approach to optimize neurocontrollers for quadrotor landing using optical flow, demonstrating that abstraction and piece-wise linear control improve real-world robustness despite sensor differences.

Contribution

It introduces a novel neurocontroller optimization method that enhances robustness and transferability of landing policies from simulation to real-world MAVs.

Findings

Piece-wise linear control outperforms simple linear schemes.

Input/output abstraction improves robustness to reality gap.

Optimized policies transfer effectively across different sensors.

Abstract

Automatic optimization of robotic behavior has been the long-standing goal of Evolutionary Robotics. Allowing the problem at hand to be solved by automation often leads to novel approaches and new insights. A common problem encountered with this approach is that when this optimization occurs in a simulated environment, the optimized policies are subject to the reality gap when implemented in the real world. This often results in sub-optimal behavior, if it works at all. This paper investigates the automatic optimization of neurocontrollers to perform quick but safe landing maneuvers for a quadrotor micro air vehicle using the divergence of the optical flow field of a downward looking camera. The optimized policies showed that a piece-wise linear control scheme is more effective than the simple linear scheme commonly used, something not yet considered by human designers. Additionally, we show the utility in using abstraction on the input and output of the controller as a tool to improve the robustness of the optimized policies to the reality gap by testing our policies optimized in simulation on real world vehicles. We tested the neurocontrollers using two different methods to generate and process the visual input, one using a conventional CMOS camera and one a dynamic vision sensor, both of which perform significantly differently than the simulated sensor. The use of the abstracted input resulted in near seamless transfer to the real world with the controllers showing high robustness to a clear reality gap.