Litcius/Paper detail

Safety-Aware Perception for Autonomous Collision Avoidance in Dynamic Environments

Ryan M. Bena, Caimeng Zhao, Quan Nguyen

2023IEEE Robotics and Automation Letters16 citationsDOIOpen Access PDF

Abstract

Autonomous collision avoidance requires accurate environmental perception; however, flight systems often possess limited sensing capabilities with field-of-view (FOV) restrictions. To navigate this challenge, we present a safety-aware approach for online determination of the optimal sensor-pointing direction <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\psi _\text{d}$</tex-math></inline-formula> which utilizes control barrier functions (CBFs). First, we generate a spatial density function <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\Phi$</tex-math></inline-formula> which leverages CBF constraints to map the collision risk of all local coordinates. Then, we convolve <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\Phi$</tex-math></inline-formula> with an attitude-dependent sensor FOV quality function to produce the objective function <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\Gamma$</tex-math></inline-formula> which quantifies the total observed risk for a given pointing direction. Finally, by finding the global optimizer for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\Gamma$</tex-math></inline-formula> , we identify the value of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\psi _\text{d}$</tex-math></inline-formula> which maximizes the perception of risk within the FOV. We incorporate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\psi _\text{d}$</tex-math></inline-formula> into a safety-critical flight architecture and conduct a numerical analysis using multiple simulated mission profiles. Our algorithm achieves a success rate of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{88}-\text{96}\%$</tex-math></inline-formula> , constituting a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{16}-\text{29}\%$</tex-math></inline-formula> improvement compared to the best heuristic methods. We demonstrate the functionality of our approach via a flight demonstration using the Crazyflie 2.1 micro-quadrotor. Without a priori obstacle knowledge, the quadrotor follows a dynamic flight path while simultaneously calculating and tracking <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\psi _\text{d}$</tex-math></inline-formula> to perceive and avoid two static obstacles with an average computation time of 371 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mu$</tex-math></inline-formula> s.

Topics & Concepts

Collision avoidancePerceptionComputer scienceCollisionHuman–computer interactionPsychologyComputer securityNeuroscienceAutonomous Vehicle Technology and SafetyRobotic Path Planning AlgorithmsReinforcement Learning in Robotics
Safety-Aware Perception for Autonomous Collision Avoidance in Dynamic Environments | Litcius