Perception-Aware Perching on Powerlines With Multirotors

Abstract

Multirotor aerial robots are becoming widely used for the inspection of powerlines. To enable continuous, robust inspection without human intervention, the robots must be able to perch on the powerlines to recharge their batteries. Highly versatile perching capabilities are necessary to adapt to the variety of configurations and constraints that are present in real powerline systems. This paper presents a novel perching trajectory generation framework that computes perception-aware, collisionfree, and dynamically-feasible maneuvers to guide the robot to the desired final state. Trajectory generation is achieved via solving a Nonlinear Programming problem using the Primal-Dual Interior Point method. The problem considers the full dynamic model of the robot down to its single rotor thrusts and minimizes the final pose and velocity errors while avoiding collisions and maximizing the visibility of the powerline during the maneuver. The generated maneuvers consider both the perching and the posterior recovery trajectories. The framework adopts costs and constraints defined by efficient mathematical representations of powerlines, enabling online onboard execution in resource-constrained hardware. The method is validated on-board an agile quadrotor conducting powerline inspection and various perching maneuvers with final pitch values of up to 180°. The developed code is available online at: https://github.com/grvcPerception/pa powerline perching

Real-Time Trajectory Planning for Aerial Perching

“…Then the robots can attach to smooth surfaces like glass or tiles, leveraging the specified terminal normal speed. There is also some work [1]- [6] focusing on the problem of perching on cylindrical objects or cables. In these methods, robots approach the targets gradually while hovering and then use claws or other similar gripers to perch on the objects.…”

“…Other methods [6,15] formulate a constrained discretetime non-linear MPC to satisfy complex constraints but cost much more computation time. Vlantis et al [15] first study the problem of landing a quadrotor on an inclined moving platform.…”

“…Moreover, the slope of the inclined surface is quite small, which is not so difficult to land on. Paneque et al [6] formulate a discrete-time multipleshooting NLP problem for perching on powerlines, which computes perception-aware, collision-free, and dynamicallyfeasible maneuvers to guide the robot to the desired final state. Notably, the generated maneuvers consider both the perching and the posterior recovery trajectories.…”

“…We benchmark our proposed method with an open-source state-of-the-art perching method [6]. This work formulates a constrained discrete-time NLP problem and solves it using ForcesPRO [22].…”

“…For a perfect perching flight, the relative tangential velocity to the surface at the landing moment should be zero to avoid scratching the surface. Most existing perching systems set a fixed desired terminal velocity according to their variant perching mechanism: zero for grippers [1]- [6], a large normal velocity for suction [7,8] and adhesives [9]- [14]. Nevertheless, such constraint requires the drone to dive in advance and is prone to collisions with the ground when there is no enough space.…”