Direct Force and Pose NMPC with Multiple Interaction Modes for Aerial Push-and-Slide Operations

Abstract

In this paper, we present a model predictive controller for a fully actuated aerial manipulator to track a hybrid force and pose trajectory at the end-effector in an aerial interaction task. A force sensor at the end-effector is used to detect contact and to directly control the interaction force. We propose an approach for automatic transition between three operation modes which reflect the state of contact constraints, including free flight and two modes for force control based on static or dynamic friction at the end-effector. This division into three modes allows for different mode-specific controller tunings to optimize the desired performance throughout an interaction task. Results from flight experiments which combine force, position, and attitude tracking, show the performance of the controller in terms of accuracy and precision. The performance is further benchmarked against a hybrid force/impedance controller.

Three Fundamental Paradigms for Aerial Physical Interaction Using Nonlinear Model Predictive Control

“…The position will be controlled in the unconstrained DoF , while the force will be controlled in the constrained DoF [15]. Similar to [14], two different control modes will be defined, namely, Free-flight Mode (FM), and Contact Mode (CM). In FM, the system's motion is not constrained in any DoF , and in CM the system's motion is constrained along at least one DoF .…”

“…Similarly, accurate aerial writing was demonstrated using a hexarotor that is equipped with a parallel arm in [13], and controlled by a hybrid force and position Nonlinear Model Predictive Control (NMPC). Recently, [14] proposed an NMPC to directly control the force and pose of an AR during push-and-slide tasks. The control approach uses a smooth switching strategy between 3 different modes of control, namely free-flight, static, and dynamic interaction.…”

MPC with Learned Residual Dynamics with Application on Omnidirectional MAVs

“…More recently, modelbased optimal control approaches have been studied as well. Model predictive control (MPC) can be specifically useful in situations where accurate and fast trajectory tracking is needed in the presence of actuator constraints and external disturbances [11]. In this context, [12] uses a MPC on an underactuated quadrotor in strong wind gusts, comparing different Kalman filters for disturbance estimation.…”

Drone-assisted collection of environmental DNA from tree branches for biodiversity monitoring

“…The requirement of touching branches to collect eDNA translates into the need for the drone to establish contact with the surroundings by applying forces on surfaces. Aerial physical interaction requires a combination of hardware (e.g., omnidirectional vehicles, protective structures, robotic arms, and end effectors) (33), direct (34,35) or indirect (36,37) force sensing, and control strategies including the most popular impedance and admittance control (38)(39)(40). Interaction tasks, such as perching (41,42), collision handling (43,44), contact-based inspection (45)(46)(47), and aerial manipulation (48,49), are typically limited to structures with rigid surfaces.…”