Learning Variable Impedance Control for Aerial Sliding on Uneven Heterogeneous Surfaces by Proprioceptive and Tactile Sensing

Abstract

The recent development of novel aerial vehicles capable of physically interacting with the environment leads to new applications such as contact-based inspection. These tasks require the robotic system to exchange forces with partially-known environments, which may contain uncertainties including unknown spatially-varying friction properties and discontinuous variations of the surface geometry. Finding a control strategy that is robust against these environmental uncertainties remains an open challenge. This paper presents a learning-based adaptive control strategy for aerial sliding tasks. In particular, the gains of a standard impedance controller are adjusted in real-time by a policy based on the current control signals, proprioceptive measurements, and tactile sensing. This policy is trained in simulation with simplified actuator dynamics in a student-teacher learning setup. The real-world performance of the proposed approach is verified using a tiltarm omnidirectional flying vehicle. The proposed controller structure combines data-driven and model-based control methods, enabling our approach to successfully transfer directly and without adaptation from simulation to the real platform. Compared to fine-tuned state of the art interaction control methods we achieve reduced tracking error and improved disturbance rejection.

Full-Body Control of an Aerial Manipulator for Advance Physical Interaction Using Fuzzy Reinforcement Learning: A Road Map

“…The flowchart of proposed method is depicted in figure 2. On the other hand, [12] went on coupling approach. The point contact robot is modeled as a single rigid body and is able to measure contact wrench data using a force/torque sensor as well as control it using an Impedance control with a gain-adjusting policy.…”

“…In [13] attitude control of a novel aircraft with low stability characteristics is addressed using Fuzzy Qlearning (FQL). Although the reward function definition in [13], and [12] are analogous, the robustness performance of the proposed FQL was outstanding against actuator faults, atmospheric disturbances, model parameter uncertainties, and sensor measurement errors in comparison with well-known PID and Dynamic Inversion methods. Furthermore, the FQL eliminates the problem of computational resources that are needed for Deep Reinforcement Learning algorithms.…”

Learning to Open Doors with an Aerial Manipulator

“…However, despite its potential, this approach is so far rarely adopted in the aerial manipulation field. A few examples include aerial transportation of suspended loads [25] and, very recently, push-and-slide tasks [26], but aerial manipulation of articulated objects has not yet been explored using RL.…”

Multi-directional Interaction Force Control with an Aerial Manipulator Under External Disturbances

“…1 Picture of the OMAV aerial robot equipped with a manipulator slide capabilities. Interactive push-and-slide tasks have been extensively studied first for robotic manipulators (Ortenzi et al, 2017) and lately for aerial robots (Tognon et al, 2019;Nguyen & Lee, 2013;Park et al, 2018) as well, through the use of hybrid force-motion controllers (Nava et al, 2020), and learning-based adaptive strategies (Zhang et al, 2022). When AM interacts with the environment, it needs to control at the same time the position at the contact, and the interaction force, preserving the stability of the entire system (Meng et al, 2018a).…”