Gökhan Alcan
Assistant Professor
Tampere University
Publications
Abstract
Robotic magnetic manipulation offers a minimally invasive approach to gastrointestinal examinations through capsule endoscopy. However, controlling such systems using external permanent magnets (EPM) is challenging due to nonlinear magnetic interactions, especially when there are complex navigation requirements such as avoidance of sensitive tissues. In this work, we present a novel trajectory planning and control method incorporating dynamics and navigation requirements, using a single EPM fixed to a robotic arm to manipulate an internal permanent magnet (IPM). Our approach employs a constrained iterative linear quadratic regulator that considers the dynamics of the IPM to generate optimal trajectories for both the EPM and IPM. Extensive simulations and real-world experiments, motivated by capsule endoscopy operations, demonstrate the robustness of the method, showcasing resilience to external disturbances and precise control under varying conditions. The experimental results show that the IPM reaches the goal position with a maximum mean error of 0.18 cm and a standard deviation of 0.21 cm. This work introduces a unified framework for constrained trajectory optimization in magnetic manipulation, directly incorporating both the IPM’s dynamics and the EPM’s manipulability.
Abstract
Hybrid dynamical systems pose significant challenges for effective planning and control, especially when additional constraints such as obstacle avoidance, state boundaries, and actuation limits are present. In this letter, we extend the recently proposed Hybrid iLQR method [1] to handle state and input constraints within an indirect optimization framework, aiming to preserve computational efficiency and ensure dynamic feasibility. Specifically, we incorporate two constraint handling mechanisms into the Hybrid iLQR: Discrete Barrier State and Augmented Lagrangian methods. Comprehensive simulations across various operational situations are conducted to evaluate and compare the performance of these extended methods in terms of convergence and their ability to handle infeasible starting trajectories. Results indicate that while the Discrete Barrier State approach is more computationally efficient, the Augmented Lagrangian method outperforms it in complex and real-world scenarios with infeasible initial trajectories.
Learning Transparent Reward Models via Unsupervised Feature Selection
In complex real-world tasks such as robotic manipulation and autonomous driving, collecting expert demonstrations is often more straightforward than specifying precise learning objectives and task descriptions. Learning from expert data can be achieved through behavioral cloning or by learning a reward function, i.e., inverse reinforcement learning. The latter allows for training with additional data outside the training distribution, guided by the inferred reward function. We propose a novel approach to construct compact and transparent reward models from automatically selected state features. These inferred rewards have an explicit form and enable the learning of policies that closely match expert behavior by training standard reinforcement learning algorithms from scratch. We validate our method’s performance in various robotic environments with continuous and high-dimensional state spaces.
Constrained Trajectory Optimization on Matrix Lie Groups via Lie-Algebraic Differential Dynamic Programming
Abstract
Matrix Lie groups are an important class of manifolds commonly used in control and robotics, and optimizing control policies on these manifolds is a fundamental problem. In this work, we propose a novel and computationally efficient approach for trajectory optimization on matrix Lie groups using an augmented Lagrangian-based constrained discrete Differential Dynamic Programming (DDP). The method involves lifting the optimization problem to the Lie algebra during the backward pass and retracting back to the manifold during the forward pass. Unlike previous approaches that addressed constraint handling only for specific classes of matrix Lie groups, the proposed method provides a general solution for nonlinear constraint handling across generic matrix Lie groups. We evaluate the effectiveness of the proposed DDP method in handling constraints within a mechanical system characterized by rigid body dynamics in SE(3), assessing its computational efficiency compared to existing direct optimization solvers. Additionally, the method demonstrates robustness under external disturbances when applied as a Lie-algebraic feedback control policy on SE(3), and in optimizing a quadrotor’s trajectory in a challenging realistic scenario. Experiments show that the proposed approach effectively manages general constraints defined on configuration, velocity, and inputs during optimization, while also maintaining stability under external disturbances when executing the resultant control policy in closed-loop.
Keywords
- Matrix Lie Groups
- Geometric Control
- Trajectory Optimization
- Constrained Optimization
- Differential Dynamic Programming
BibTeX
@article{alcan2024cddplie,
title={Constrained Trajectory Optimization on Matrix Lie Groups via Lie-Algebraic Differential Dynamic Programming},
author={Alcan, Gokhan and Abu-Dakka, Fares J and Kyrki, Ville},
journal={arXiv preprint arXiv:2301.02018},
year={2024}
}
Data-driven Diffusion Models for Enhancing Safety in Autonomous Vehicle Traffic Simulations
Safety-critical traffic scenarios are integral to the development and validation of autonomous driving systems. These scenarios provide crucial insights into vehicle responses under high-risk conditions rarely encountered in real-world settings. Recent advancements in critical scenario generation have demonstrated the superiority of diffusion-based approaches over traditional generative models in terms of effectiveness and realism. However, current diffusion-based methods fail to adequately address the complexity of driver behavior and traffic density information, both of which significantly influence driver decision-making processes. In this work, we present a novel approach to overcome these limitations by introducing adversarial guidance functions for diffusion models that incorporate behavior complexity and traffic density, thereby enhancing the generation of more effective and realistic safety-critical traffic scenarios. The proposed method is evaluated on two evaluation metrics: effectiveness and realism.
Trajectory Optimization on Matrix Lie Groups with Differential Dynamic Programming and Nonlinear Constraints
Matrix Lie groups are an important class of manifolds commonly used in control and robotics, and the optimization of control policies on these manifolds is a fundamental problem. In this work, we propose a new approach for trajectory optimization on matrix Lie groups using an augmented Lagrangian-based constrained Differential Dynamic Programming (DDP). The method involves lifting the optimization problem to the Lie algebra in the backward pass and retracting back to the manifold in the forward pass. In contrast to previous approaches which only addressed constraint handling for specific classes of matrix Lie groups, the proposed method provides a general approach for nonlinear constraint handling for generic matrix Lie groups. We evaluate the effectiveness of the proposed DDP method in handling constraints in a simple mechanical system characterized by rigid body dynamics in SE(3), assessing computational efficiency compared to existing optimization solvers, and optimizing trajectory performance in a realistic quadrotor scenario.
Automated Feature Selection for Inverse Reinforcement Learning
Abstract
Inverse reinforcement learning (IRL) is an imitation learning approach to learning reward functions from expert demonstrations. Its use avoids the difficult and tedious procedure of manual reward specification while retaining the generalization power of reinforcement learning. In IRL, the reward is usually represented as a linear combination of features. In continuous state spaces, the state variables alone are not sufficiently rich to be used as features, but which features are good is not known in general. To address this issue, we propose a method that employs polynomial basis functions to form a candidate set of features, which are shown to allow the matching of statistical moments of state distributions. Feature selection is then performed for the candidates by leveraging the correlation between trajectory probabilities and feature expectations. We demonstrate the approach’s effectiveness by recovering reward functions that capture expert policies across non-linear control tasks of increasing complexity.
Benchmarking the Sim-to-Real Gap in Cloth Manipulation
Abstract
Realistic physics engines play a crucial role for learning to manipulate deformable objects such as garments in simulation. By doing so, researchers can circumvent challenges such as sensing the deformation of the object in the real-world. In spite of the extensive use of simulations for this task, few works have evaluated the reality gap between deformable object simulators and real-world data. We present a benchmark dataset to evaluate the sim-to-real gap in cloth manipulation. The dataset is collected by performing a dynamic cloth manipulation task involving contact with a rigid table. We use the dataset to evaluate the reality gap, computational time, and simulation stability of four popular deformable object simulators: MuJoCo, Bullet, Flex, and SOFA. Additionally, we discuss the benefits and drawbacks of each simulator.
Keywords
- Datasets for Robot Learning
- Bimanual Manipulation
- Deformable Object Manipulation
BibTeX
@article{blanco2024benchmarking,
title={Benchmarking the sim-to-real gap in cloth manipulation},
author={Blanco-Mulero, David and Barbany, Oriol and Alcan, Gokhan and Colom{\'e}, Adri{\`a} and Torras, Carme and Kyrki, Ville},
journal={IEEE Robotics and Automation Letters},
volume={9},
number={3},
pages={1760--1767},
year={2024},
publisher={IEEE}
}
The Role of Higher-Order Cognitive Models in Active Learning
Building machines capable of efficiently collaborating with humans has been a longstanding goal in artificial intelligence. Especially in the presence of uncertainties, optimal cooperation often requires that humans and artificial agents model each other’s behavior and use these models to infer underlying goals, beliefs or intentions, potentially involving multiple levels of recursion. Empirical evidence for such higher-order cognition in human behavior is also provided by previous works in cognitive science, linguistics, and robotics. We advocate for a new paradigm for active learning for human feedback that utilises humans as active data sources while accounting for their higher levels of agency. In particular, we discuss how increasing level of agency results in qualitatively different forms of rational communication between an active learning system and a teacher. Additionally, we provide a practical example of active learning using a higher-order cognitive model. This is accompanied by a computational study that underscores the unique behaviors that this model produces.
Challenges of Data-Driven Simulation of Diverse and Consistent Human Driving Behaviors
Building simulation environments for developing and testing autonomous vehicles necessitates that the simulators accurately model the statistical realism of the real-world environment, including the interaction with other vehicles driven by human drivers. To address this requirement, an accurate human behavior model is essential to incorporate the diversity and consistency of human driving behavior. We propose a mathematical framework for designing a data-driven simulation model that simulates human driving behavior more realistically than the currently used physics-based simulation models. Experiments conducted using the NGSIM dataset validate our hypothesis regarding the necessity of considering the complexity, diversity, and consistency of human driving behavior when aiming to develop realistic simulators.