Learning-based Nonlinear Model Predictive Control of Articulated Soft Robots using Recurrent Neural Networks

Soft robots pose difficulties in terms of control, requiring novel strategies to effectively manipulate their compliant structures. Model-based approaches face challenges due to the high dimensionality and nonlinearities such as hysteresis effects. In contrast, learning-based approaches provide nonlinear models of different soft robots based only on measured data. In this paper, recurrent neural networks (RNNs) predict the behavior of an articulated soft robot (ASR) with five degrees of freedom (DoF). RNNs based on gated recurrent units (GRUs) are compared to the more commonly used long short-term memory (LSTM) networks and show better accuracy. The recurrence enables the capture of hysteresis effects that are inherent in soft robots due to viscoelasticity or friction but cannot be captured by simple feedforward networks. The data-driven model is used within a nonlinear model predictive control (NMPC), whereby the correct handling of the RNN's hidden states is focused. A training approach is presented that allows measured values to be utilized in each control cycle. This enables accurate predictions of short horizons based on sensor data, which is crucial for closed-loop NMPC. The proposed learning-based NMPC enables trajectory tracking with an average error of 1.2deg in experiments with the pneumatic five-DoF ASR.

Autonomous Iterative Motion Learning (AI-MOLE) of a SCARA Robot for Automated Myocardial Injection

Stem cell therapy is a promising approach to treat heart insufficiency and benefits from automated myocardial injection which requires highly precise motion of a robotic manipulator that is equipped with a syringe. This work investigates whether sufficiently precise motion can be achieved by combining a SCARA robot and learning control methods. For this purpose, the method Autonomous Iterative Motion Learning (AI-MOLE) is extended to be applicable to multi-input/multi-output systems. The proposed learning method solves reference tracking tasks in systems with unknown, nonlinear, multi-input/multi-output dynamics by iteratively updating an input trajectory in a plug-and-play fashion and without requiring manual parameter tuning. The proposed learning method is validated in a preliminary simulation study of a simplified SCARA robot that has to perform three desired motions. The results demonstrate that the proposed learning method achieves highly precise reference tracking without requiring any a priori model information or manual parameter tuning in as little as 15 trials per motion. The results further indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.

Dispelling Four Challenges in Inertial Motion Tracking with One Recurrent Inertial Graph-based Estimator (RING)

In this paper, we extend the Recurrent Inertial Graph-based Estimator (RING), a novel neural-network-based solution for Inertial Motion Tracking (IMT), to generalize across a large range of sampling rates, and we demonstrate that it can overcome four real-world challenges: inhomogeneous magnetic fields, sensor-to-segment misalignment, sparse sensor setups, and nonrigid sensor attachment. RING can estimate the rotational state of a three-segment kinematic chain with double hinge joints from inertial data, and achieves an experimental mean-absolute-(tracking)-error of 8.10 +/- 1.19 degrees if all four challenges are present simultaneously. The network is trained on simulated data yet evaluated on experimental data, highlighting its remarkable ability to zero-shot generalize from simulation to experiment. We conduct an ablation study to analyze the impact of each of the four challenges on RING's performance, we showcase its robustness to varying sampling rates, and we demonstrate that RING is capable of real-time operation. This research not only advances IMT technology by making it more accessible and versatile but also enhances its potential for new application domains including non-expert use of sparse IMT with nonrigid sensor attachments in unconstrained environments.

Towards Optimized Parallel Robots for Human-Robot Collaboration by Combined Structural and Dimensional Synthesis

Parallel robots (PR) offer potential for human-robot collaboration (HRC) due to their lower moving masses and higher speeds. However, the parallel leg chains increase the risks of collision and clamping. In this work, these hazards are described by kinematics and kinetostatics models to minimize them as objective functions by a combined structural and dimensional synthesis in a particle-swarm optimization. In addition to the risk of clamping within and between kinematic chains, the back-drivability is quantified to theoretically guarantee detectability via motor current. Another HRC-relevant objective function is the largest eigenvalue of the mass matrix formulated in the operational-space coordinates to consider collision effects. Multi-objective optimization leads to different Pareto-optimal PR structures. The results show that the optimization leads to significant improvement of the HRC criteria and that a Hexa structure (6-RUS) is to be favored concerning the objective functions and due to its simpler joint structure.

Domain-decoupled Physics-informed Neural Networks with Closed-form Gradients for Fast Model Learning of Dynamical Systems

Physics-informed neural networks (PINNs) are trained using physical equations and can also incorporate unmodeled effects by learning from data. PINNs for control (PINCs) of dynamical systems are gaining interest due to their prediction speed compared to classical numerical integration methods for nonlinear state-space models, making them suitable for real-time control applications. We introduce the domain-decoupled physics-informed neural network (DD-PINN) to address current limitations of PINC in handling large and complex nonlinear dynamic systems. The time domain is decoupled from the feed-forward neural network to construct an Ansatz function, allowing for calculation of gradients in closed form. This approach significantly reduces training times, especially for large dynamical systems, compared to PINC, which relies on graph-based automatic differentiation. Additionally, the DD-PINN inherently fulfills the initial condition and supports higher-order excitation inputs, simplifying the training process and enabling improved prediction accuracy. Validation on three systems - a nonlinear mass-spring-damper, a five-mass-chain, and a two-link robot - demonstrates that the DD-PINN achieves significantly shorter training times. In cases where the PINC's prediction diverges, the DD-PINN's prediction remains stable and accurate due to higher physics loss reduction or use of a higher-order excitation input. The DD-PINN allows for fast and accurate learning of large dynamical systems previously out of reach for the PINC.

A Soft Robotic System Automatically Learns Precise Agile Motions Without Model Information

Many application domains, e.g., in medicine and manufacturing, can greatly benefit from pneumatic Soft Robots (SRs). However, the accurate control of SRs has remained a significant challenge to date, mainly due to their nonlinear dynamics and viscoelastic material properties. Conventional control design methods often rely on either complex system modeling or time-intensive manual tuning, both of which require significant amounts of human expertise and thus limit their practicality. In recent works, the data-driven method, Automatic Neural ODE Control (ANODEC) has been successfully used to -- fully automatically and utilizing only input-output data -- design controllers for various nonlinear systems in silico, and without requiring prior model knowledge or extensive manual tuning. In this work, we successfully apply ANODEC to automatically learn to perform agile, non-repetitive reference tracking motion tasks in a real-world SR and within a finite time horizon. To the best of the authors' knowledge, ANODEC achieves, for the first time, performant control of a SR with hysteresis effects from only 30 seconds of input-output data and without any prior model knowledge. We show that for multiple, qualitatively different and even out-of-training-distribution reference signals, a single feedback controller designed by ANODEC outperforms a manually tuned PID baseline consistently. Overall, this contribution not only further strengthens the validity of ANODEC, but it marks an important step towards more practical, easy-to-use SRs that can automatically learn to perform agile motions from minimal experimental interaction time.

SPONGE: Open-Source Designs of Modular Articulated Soft Robots

Soft-robot designs are manifold, but only a few are publicly available. Often, these are only briefly described in their publications. This complicates reproduction, and hinders the reproducibility and comparability of research results. If the designs were uniform and open source, validating researched methods on real benchmark systems would be possible. To address this, we present two variants of a soft pneumatic robot with antagonistic bellows as open source. Starting from a semi-modular design with multiple cables and tubes routed through the robot body, the transition to a fully modular robot with integrated microvalves and serial communication is highlighted. Modularity in terms of stackability, actuation, and communication is achieved, which is the crucial requirement for building soft robots with many degrees of freedom and high dexterity for real-world tasks. Both systems are compared regarding their respective advantages and disadvantages. The robots' functionality is demonstrated in experiments on airtightness, gravitational influence, position control with mean tracking errors of <3 deg, and long-term operation of cast and printed bellows. All soft- and hardware files required for reproduction are provided.

AI-MOLE: Autonomous Iterative Motion Learning for Unknown Nonlinear Dynamics with Extensive Experimental Validation

This work proposes Autonomous Iterative Motion Learning (AI-MOLE), a method that enables systems with unknown, nonlinear dynamics to autonomously learn to solve reference tracking tasks. The method iteratively applies an input trajectory to the unknown dynamics, trains a Gaussian process model based on the experimental data, and utilizes the model to update the input trajectory until desired tracking performance is achieved. Unlike existing approaches, the proposed method determines necessary parameters automatically, i.e., AI-MOLE works plug-and-play and without manual parameter tuning. Furthermore, AI-MOLE only requires input/output information, but can also exploit available state information to accelerate learning. While other approaches are typically only validated in simulation or on a single real-world testbed using manually tuned parameters, we present the unprecedented result of validating the proposed method on three different real-world robots and a total of nine different reference tracking tasks without requiring any a priori model information or manual parameter tuning. Over all systems and tasks, AI-MOLE rapidly learns to track the references without requiring any manual parameter tuning at all, even if only input/output information is available.

Quantifying Uncertainties of Contact Classifications in a Human-Robot Collaboration with Parallel Robots

In human-robot collaboration, unintentional physical contacts occur in the form of collisions and clamping, which must be detected and classified separately for a reaction. If certain collision or clamping situations are misclassified, reactions might occur that make the true contact case more dangerous. This work analyzes data-driven modeling based on physically modeled features like estimated external forces for clamping and collision classification with a real parallel robot. The prediction reliability of a feedforward neural network is investigated. Quantification of the classification uncertainty enables the distinction between safe versus unreliable classifications and optimal reactions like a retraction movement for collisions, structure opening for the clamping joint, and a fallback reaction in the form of a zero-g mode. This hypothesis is tested with experimental data of clamping and collision cases by analyzing dangerous misclassifications and then reducing them by the proposed uncertainty quantification. Finally, it is investigated how the approach of this work influences correctly classified clamping and collision scenarios.

VQF: Highly Accurate IMU Orientation Estimation with Bias Estimation and Magnetic Disturbance Rejection

The miniaturization of inertial measurement units (IMUs) facilitates their widespread use in a growing number of application domains. Orientation estimation is a prerequisite for most further data processing steps in inertial motion tracking, such as position/velocity estimation, joint angle estimation, and 3D visualization. Errors in the estimated orientations severely affect all further processing steps. Few existing publications systematically compare multiple algorithms on a broad collection of experimental data, and those publications show that out-of-the-box accuracy of existing algorithms is often low and that application-specific tuning is required. In the present work, we propose and extensively evaluate an orientation estimation algorithm that is based on a novel approach of filtering the acceleration measurements in an almost-inertial frame and that includes extensions for gyroscope bias estimation and magnetic disturbance rejection, as well as a variant for offline data processing. In contrast to all existing work, we perform a comprehensive evaluation, using a large collection of publicly available datasets and eight literature methods for comparison. The proposed method consistently outperforms all literature methods and achieves an average RMSE of 2.9{\deg}, while the errors obtained with literature methods range from 5.3{\deg} to 16.7{\deg}. Since the evaluation was performed with one single fixed parametrization across a very diverse dataset collection, we conclude that the proposed method provides unprecedented out-of-the-box performance for a broad range of motions, sensor hardware, and environmental conditions. This gain in orientation estimation accuracy is expected to advance the field of IMU-based motion analysis and provide performance benefits in numerous applications. The provided open-source implementation makes it easy to employ the proposed method.