Efficient trajectory optimization for generalized multirotors via sequential convex programming and convexity exploitation
Abstract
Aerial interaction requires aerial robots capable of independent attitude control, leading to the rising popularity of omnidirectional tiltable multirotors (tilt-multirotors). Unlike standard multirotors that rely on differential flatness to generate high-quality trajectories, tilt-multirotors possess servo-integrated nonlinear dynamics. Consequently, their trajectory optimization becomes a challenging nonconvex problem, generally suffering from high initialization sensitivity and computational costs. To address these challenges, we propose the penalized trust-region sequential convex programming with line search framework.
Type
Publication
IEEE Transactions on Automation Science and Engineering
