Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Peter Manzl has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Path following and stabilization of a bicycle model using a reinforcement learning approach

    by Sebastian Weyrer, Peter Manzl, A. L. Schwab, Johannes Gerstmayr

    Over the years, complex control approaches have been developed to control the motion of a bicycle. Reinforcement Learning (RL), a branch of machine learning, promises to be an automated approach for solving optimal control problems. By interacting with and observing an environment, a so-called agent is trained, ultimately leading to a learned controller. The present work introduces a pure RL approach to do path following with a virtual bicycle model while simultaneously stabilizing it laterally. The bicycle, modeled using the Whipple benchmark model and multibody system dynamics, has no stabilization aids. The observation of the environment consists of the minimal positional and velocity coordinates of the bicycle, as well as of information about the path ahead of the bicycle provided by moving preview points. Both path following and stabilization of the bicycle model are achieved exclusively by controlling the steering angle setpoint of the bicycle. Curriculum learning is applied as a state-of-the-art training strategy. Different settings for the RL approach are investigated and compared. The ability of the learned controllers to do path following and stabilization of the bicycle model traveling between 2 m/s and 7 m/s along complex paths including full circles, slalom maneuvers, and lane changes is demonstrated. Explanatory methods for machine learning are used to analyze the learned controller and identify connections to research in bicycle dynamics.


  • Open Access

    Real-time structural dynamics estimation in hydraulically actuated systems using 3D flexible multibody simulation and SLIDE networks

    by Qasim Khadim, Peter Manzl, Emil Kurvinen, Aki Mikkola, Grzegorz Orzechowski, Johannes Gerstmayr


  • Surrogate Models for Compliant Joints in Programmable Structures

    by Michael Pieber, Zhaowei Zhang, Peter Manzl, Johannes Gerstmayr

    This work focuses on compliant joints in autonomous cells. These cells preserve the underlying geometry of a triangular mesh and enable self-reconfiguration using six-bar linkages. The six-bar linkages, essential for maintaining mesh geometry, are realized as 3D-printable compliant mechanisms. However, compliance in the hinges and deviations from the desired remote center of rotation results in positioning errors. Detailed multibody models have been developed using nonlinear beam elements to accurately represent the compliant mechanisms. In order to meet the required computational real-time performance, we propose surrogate models that can accurately predict positioning errors during self-reconfiguration. Finally, these errors are corrected by solving the inverse kinematics of a hyper-redundant manipulator using the static solution of a linearized model. Our study illustrates the benefits of utilizing a surrogate model, which reduces CPU time compared to beam elements. For the first time, we have successfully corrected positioning errors within a system of cells.


  • Open Access

    Creation, Evaluation and Self-Validation of Simulation Models with Large Language Models

    by Tobias Möltner, Peter Manzl, Michael Pieber, Johannes Gerstmayr

    Engineering tasks are significantly underrepresented in current large language model (LLM) datasets and research, despite their complexity and practical importance. These tasks often demand a deep mathematical and involve a combination of textual descriptions, visual representations, and numerical data. Moreover, engineering frequently relies on accepted approximations and models rather than exact values. Therefore, the present paper advances the integration of LLMs into mechanical engineering by introducing a comprehensive framework for automated simulation model generation and validation. The framework is designed as a benchmark and focuses on mechanical engineering problems in dynamics in particular on multibody dynamics simulation models in Python. It allows for the creation of a large number of test cases due to its use of parametrized models with ground truth solutions, allowing evaluation for executability and correctness. Lastly, LLM-agents are employed to generate simulation models and perform self-evaluation through a predefined set of validation methods, assessing models for parametrization errors. Evaluation results using classical F-score metrics demonstrate that most tested LLMs identify a majority of incorrect models, while the best-performing model achieves high accuracy in differing between correct and wrong simulation models.


  • Open Access

    SLIDE: A machine-learning based method for forced dynamic response estimation of multibody systems

    by Peter Manzl, Alexander Humer, Qasim Khadim, Johannes Gerstmayr

    In computational engineering, enhancing the simulation speed and efficiency is a perpetual goal. To fully take advantage of neural network techniques and hardware, we present the SLiding-window Initially-truncated Dynamic-response Estimator (SLIDE), a deep learning-based method designed to estimate output sequences of mechanical or multibody systems with primarily, but not exclusively, forced excitation. A key advantage of SLIDE is its ability to estimate the dynamic response of damped systems without requiring the full system state, making it particularly effective for flexible multibody systems. The method truncates the output window based on the decay of initial effects, such as damping, which is approximated by the complex eigenvalues of the systems linearized equations. In addition, a second neural network is trained to provide an error estimation, further enhancing the methods applicability. The method is applied to a diverse selection of systems, including the Duffing oscillator, a flexible slider-crank system, and an industrial 6R manipulator, mounted on a flexible socket. Our results demonstrate significant speedups from the simulation up to several millions, exceeding real-time performance substantially.

  • Open Access

    Task-Dependent Comfort Zone, a Base Placement Strategy for Mobile Manipulators Based on Manipulability Measures

    by Martin Sereinig, Peter Manzl, Johannes Gerstmayr

    The present contribution introduces the task-dependent comfort zone as a base placement strategy for mobile manipulators using different manipulability measures. Four different manipulability measures depending on end-effector velocities, forces, stiffness, and accelerations are considered. By evaluating a discrete subspace of the manipulator workspace with these manipulability measures and using image-processing algorithms, a suitable goal position for the autonomous mobile manipulator was defined within the comfort zone. This always ensures a certain manipulator manipulablity value with a lower limit with respect to the maximum possible manipulability in the discrete subspace. Results are shown for three different mobile manipulators using the velocity-dependent manipulability measure in a simulation.


  • Open Access

    Task Dependent Comfort Zone, a Base Placement Strategy for Mobile Manipulators based on Manipulability Measures

    by Martin Sereinig, Peter Manzl, Johannes Gerstmayr

    The present contribution introduces the task dependent comfort zone as a base placement strategy for mobile manipulators using different manipulability measures. Four different manipulability measures depending on end-effector velocities, forces, stiffness, and accelerations are considered. By evaluating a discrete subspace of the manipulator workspace with these manipulability measures and by using image processing algorithms, a suitable goal position for the autonomous mobile manipulator was defined within the comfort zone. This always ensures a certain manipulator manipulablity value with a lower limit in respect to the maximum possible manipulability in the discrete subspace. Results are shown for three different mobile manipulators using the velocity dependent manipulability measure in a simulation.


  • Open Access

    Reliability evaluation of reinforcement learning methods for mechanical systems with increasing complexity

    by Peter Manzl, Oleg Rogov, Johannes Gerstmayr, Aki Mikkola, Grzegorz Orzechowski

    Reinforcement learning (RL) is one of the emerging fields of artificial intelligence (AI) intended for designing agents that take actions in the physical environment. RL has many vital applications, including robotics and autonomous vehicles. The key characteristic of RL is its ability to learn from experience without requiring direct programming or supervision. To learn, an agent interacts with an environment by acting and observing the resulting states and rewards. In most practical applications, an environment is implemented as a virtual system due to cost, time, and safety concerns. Simultaneously, multibody system dynamics (MSD) is a framework for efficiently and systematically developing virtual systems of arbitrary complexity. MSD is commonly used to create virtual models of robots, vehicles, machinery, and humans. The features of RL and MSD make them perfect companions in building sophisticated, automated, and autonomous mechatronic systems. The research demonstrates the use of RL in controlling multibody systems. While AI methods are used to solve some of the most challenging tasks in engineering, their proper understanding and implementation are demanding. Therefore, we introduce and detail three commonly used RL algorithms to control the inverted N-pendulum on the cart. Single-, double-, and triple-pendulum configurations are investigated, showing the capability of RL methods to handle increasingly complex dynamical systems. We show 2D state space zones where the agent succeeds or fails the stabilization. Despite passing randomized tests during training, blind spots may occur where the agent’s policy fails. Results confirm that RL is a versatile, although complex, control engineering approach.


  • Open Access

    Reliability Evaluation of Reinforcement Learning Methods for Mechanical Systems with Increasing Complexity

    by Peter Manzl, Oleg Rogov, Johannes Gerstmayr, Aki Mikkola, Grzegorz Orzechowski

    Reinforcement Learning (RL) is one of the emerging fields of Artificial Intelligence (AI) intended for designing agents that take actions in the physical environment. RL has many vital applications, including robotics and autonomous vehicles. The key characteristic of RL is its ability to learn from experience without requiring direct programming or supervision. To learn, an agent interacts with an environment by acting and observing the resulting states and rewards. In most practical applications, an environment is implemented as a virtual system due to cost, time, and safety concerns. Simultaneously, Multibody System Dynamics (MSD) is a framework for efficiently and systematically developing virtual systems of arbitrary complexity. MSD is commonly used to create virtual models of robots, vehicles, machinery, and humans. The features of RL and MSD make them perfect companions in building sophisticated, automated, and autonomous mechatronic systems. The research demonstrates the use of RL in controlling multibody systems. While AI methods are used to solve some of the most challenging tasks in engineering, their proper understanding and implementation are demanding. Therefore, we introduce and detail three commonly used RL algorithms to control the inverted N-pendulum on the cart. Single-, double-, and triple-pendulum configurations are investigated, showing the capability of RL methods to handle increasingly complex dynamical systems. We show 2D state space zones where the agent succeeds or fails the stabilization. Despite passing randomized tests during training, "blind spots" may occur where the agent's policy fails. Results confirm that RL is a versatile, although complex, control engineering approach.