Trajectory Planning and Tracking of Hybrid Flying-Crawling Quadrotors

TL;DR

The paper tackles the challenge of autonomous navigation for Hybrid Flying-Crawling Quadrotors by introducing a crawling-aware terrestrial-aerial trajectory planner and a layered tracking framework that gracefully handles deformation time during mode transitions. The planner leverages a Fast-Planner–style hybrid A* with crawling-constrained terrestrial primitives and unconstrained aerial primitives, followed by trajectory optimization to ensure dynamic feasibility. A dedicated terrestrial-tracking controller with yaw-rate regulation and adaptive throttle, together with an aerial tracking loop and explicit transition logic, enables reliable switching between crawling and flying while re-planning as needed. Experimental validation on a HyFCQ platform demonstrates improved trajectory feasibility, reduced tracking errors, and successful hybrid navigation in constrained environments, highlighting the practical viability of the approach. Future work aims to incorporate structural-deformation constraints directly into the planning to further enhance speed and smoothness of HyFCQ motion.

Abstract

Hybrid Flying-Crawling Quadrotors (HyFCQs) are transformable robots with the ability of terrestrial and aerial hybrid motion. This article presents a trajectory planning and tracking framework designed for HyFCQs. In this framework, a terrestrial-aerial path-searching method with the crawling limitation of HyFCQs is proposed to guarantee the dynamical feasibility of trajectories. Additionally, a trajectory tracking method is proposed to address the challenges associated with the deformation time required by HyFCQs, which makes tracking hybrid trajectories at the junction between terrestrial and aerial segments difficult. Simulations and real-world experiments in diverse scenarios validate the exceptional performance of the proposed approach.

Figures (10)

• Figure 1: Mechanical structure diagram of the HyFCQ. (a) The body core is equipped with sensing, computing, and control units. The arms are equipped with actuation components consisting of reduction gears, timing belts, brushless motors, and wheels. They are linked to the servo mounted in the core. (b) The reduction gears transfer the torsion from the active motors to the rear wheels. The rear wheels convey the torque to the front wheels via timing belts. Symmetrical actuation of the sprawl angles on both sides is achieved through the servo. The experimental video is available at: https://youtu.be/nxFqLxel4c0
• Figure 2: Top-down view of the primitives expansion. (a) shows the process of terrestrial primitives expansion. By increasing the minimum value of $x_B$, the variation range of the yaw angle is restricted, thereby reducing the fluctuation of the yaw angle. (b) shows the process of aerial primitives expansion. Since the flying motion is unaffected by nonholonomic constraints, the primitives can expand in all directions.
• Figure 3: (a) The trajectory tracking controller for crawling mode. (b) The trajectory tracking controller for flying mode.
• Figure 4: The detailed components of the HyFCQ. In flying mode, the size of the quadrotor is $34\times 32\times 11$ cm, while in crawling mode, the size is $34\times 26\times 11$ cm. The total weight is $1.6$ kg.
• Figure 5: Top) The circular trajectory tracking. Botton) The lemniscate trajectory tracking.