arXiv subjects

Gerhard Neumann

Publications and source records attributed to Gerhard Neumann.

At least 109 records · Page 6Linked to original sources

arXiv2022

FusionVAE: A Deep Hierarchical Variational Autoencoder for RGB Image Fusion

Sensor fusion can significantly improve the performance of many computer vision tasks. However, traditional fusion approaches are either not data-driven and cannot exploit prior knowledge nor find regularities in a given dataset or they are restricted to a single application. We overcome this shortcoming by presenting a novel deep hierarchical variational autoencoder called FusionVAE that can serve as a basis for many fusion tasks. Our approach is able to generate diverse image samples that are conditioned on multiple noisy, occluded, or only partially visible input images. We derive and optimize a variational lower bound for the conditional log-likelihood of FusionVAE. In order to assess the fusion capabilities of our model thoroughly, we created three novel datasets for image fusion based on popular computer vision datasets. In our experiments, we show that FusionVAE learns a representation of aggregated information that is relevant to fusion tasks. The results demonstrate that our approach outperforms traditional methods significantly. Furthermore, we present the advantages and disadvantages of different design choices.

cs.CV↗

arXiv2022

MV6D: Multi-View 6D Pose Estimation on RGB-D Frames Using a Deep Point-wise Voting Network

Estimating 6D poses of objects is an essential computer vision task. However, most conventional approaches rely on camera data from a single perspective and therefore suffer from occlusions. We overcome this issue with our novel multi-view 6D pose estimation method called MV6D which accurately predicts the 6D poses of all objects in a cluttered scene based on RGB-D images from multiple perspectives. We base our approach on the PVN3D network that uses a single RGB-D image to predict keypoints of the target objects. We extend this approach by using a combined point cloud from multiple views and fusing the images from each view with a DenseFusion layer. In contrast to current multi-view pose detection networks such as CosyPose, our MV6D can learn the fusion of multiple perspectives in an end-to-end manner and does not require multiple prediction stages or subsequent fine tuning of the prediction. Furthermore, we present three novel photorealistic datasets of cluttered scenes with heavy occlusions. All of them contain RGB-D images from multiple perspectives and the ground truth for instance semantic segmentation and 6D pose estimation. MV6D significantly outperforms the state-of-the-art in multi-view 6D pose estimation even in cases where the camera poses are known inaccurately. Furthermore, we show that our approach is robust towards dynamic camera setups and that its accuracy increases incrementally with an increasing number of perspectives.

cs.CV↗

arXiv2022

Robot Policy Learning from Demonstration Using Advantage Weighting and Early Termination

Learning robotic tasks in the real world is still highly challenging and effective practical solutions remain to be found. Traditional methods used in this area are imitation learning and reinforcement learning, but they both have limitations when applied to real robots. Combining reinforcement learning with pre-collected demonstrations is a promising approach that can help in learning control policies to solve robotic tasks. In this paper, we propose an algorithm that uses novel techniques to leverage offline expert data using offline and online training to obtain faster convergence and improved performance. The proposed algorithm (AWET) weights the critic losses with a novel agent advantage weight to improve over the expert data. In addition, AWET makes use of an automatic early termination technique to stop and discard policy rollouts that are not similar to expert trajectories -- to prevent drifting far from the expert data. In an ablation study, AWET showed improved and promising performance when compared to state-of-the-art baselines on four standard robotic tasks.

cs.LG↗

arXiv2022

End-to-End Learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control

It is well-known that inverse dynamics models can improve tracking performance in robot control. These models need to precisely capture the robot dynamics, which consist of well-understood components, e.g., rigid body dynamics, and effects that remain challenging to capture, e.g., stick-slip friction and mechanical flexibilities. Such effects exhibit hysteresis and partial observability, rendering them, particularly challenging to model. Hence, hybrid models, which combine a physical prior with data-driven approaches are especially well-suited in this setting. We present a novel hybrid model formulation that enables us to identify fully physically consistent inertial parameters of a rigid body dynamics model which is paired with a recurrent neural network architecture, allowing us to capture unmodeled partially observable effects using the network memory. We compare our approach against state-of-the-art inverse dynamics models on a 7 degree of freedom manipulator. Using data sets obtained through an optimal experiment design approach, we study the accuracy of offline torque prediction and generalization capabilities of joint learning methods. In control experiments on the real system, we evaluate the model as a feed-forward term for impedance control and show the feedback gains can be drastically reduced to achieve a given tracking accuracy.

cs.RO↗

arXiv2022

Regret-Aware Black-Box Optimization with Natural Gradients, Trust-Regions and Entropy Control

Most successful stochastic black-box optimizers, such as CMA-ES, use rankings of the individual samples to obtain a new search distribution. Yet, the use of rankings also introduces several issues such as the underlying optimization objective is often unclear, i.e., we do not optimize the expected fitness. Further, while these algorithms typically produce a high-quality mean estimate of the search distribution, the produced samples can have poor quality as these algorithms are ignorant of the regret. Lastly, noisy fitness function evaluations may result in solutions that are highly sub-optimal on expectation. In contrast, stochastic optimizers that are motivated by policy gradients, such as the Model-based Relative Entropy Stochastic Search (MORE) algorithm, directly optimize the expected fitness function without the use of rankings. MORE can be derived by applying natural policy gradients and compatible function approximation, and is using information theoretic constraints to ensure the stability of the policy update. While MORE does not suffer from the listed limitations, it often cannot achieve state of the art performance in comparison to ranking based methods. We improve MORE by decoupling the update of the mean and covariance of the search distribution allowing for more aggressive updates on the mean while keeping the update on the covariance conservative, an improved entropy scheduling technique based on an evolution path which results in faster convergence and a simplified and more effective model learning approach in comparison to the original paper. We compare our algorithm to state of the art black-box optimization algorithms on standard optimization tasks as well as on episodic RL tasks in robotics where it is also crucial to have small regret. We obtain competitive results on benchmark functions and clearly outperform ranking-based methods in terms of regret on the RL tasks.

stat.ML↗

arXiv2022

What Matters For Meta-Learning Vision Regression Tasks?

Meta-learning is widely used in few-shot classification and function regression due to its ability to quickly adapt to unseen tasks. However, it has not yet been well explored on regression tasks with high dimensional inputs such as images. This paper makes two main contributions that help understand this barely explored area. \emph{First}, we design two new types of cross-category level vision regression tasks, namely object discovery and pose estimation of unprecedented complexity in the meta-learning domain for computer vision. To this end, we (i) exhaustively evaluate common meta-learning techniques on these tasks, and (ii) quantitatively analyze the effect of various deep learning techniques commonly used in recent meta-learning algorithms in order to strengthen the generalization capability: data augmentation, domain randomization, task augmentation and meta-regularization. Finally, we (iii) provide some insights and practical recommendations for training meta-learning algorithms on vision regression tasks. \emph{Second}, we propose the addition of functional contrastive learning (FCL) over the task representations in Conditional Neural Processes (CNPs) and train in an end-to-end fashion. The experimental results show that the results of prior work are misleading as a consequence of a poor choice of the loss function as well as too small meta-training sets. Specifically, we find that CNPs outperform MAML on most tasks without fine-tuning. Furthermore, we observe that naive task augmentation without a tailored design results in underfitting.

cs.CV↗

arXiv2022

Hierarchical Policy Learning for Mechanical Search

Retrieving objects from clutters is a complex task, which requires multiple interactions with the environment until the target object can be extracted. These interactions involve executing action primitives like grasping or pushing as well as setting priorities for the objects to manipulate and the actions to execute. Mechanical Search (MS) is a framework for object retrieval, which uses a heuristic algorithm for pushing and rule-based algorithms for high-level planning. While rule-based policies profit from human intuition in how they work, they usually perform sub-optimally in many cases. Deep reinforcement learning (RL) has shown great performance in complex tasks such as taking decisions through evaluating pixels, which makes it suitable for training policies in the context of object-retrieval. In this work, we first formulate the MS problem in a principled formulation as a hierarchical POMDP. Based on this formulation, we propose a hierarchical policy learning approach for the MS problem. For demonstration, we present two main parameterized sub-policies: a push policy and an action selection policy. When integrated into the hierarchical POMDP's policy, our proposed sub-policies increase the success rate of retrieving the target object from less than 32% to nearly 80%, while reducing the computation time for push actions from multiple seconds to less than 10 milliseconds.

cs.RO↗

arXiv2022

Specializing Versatile Skill Libraries using Local Mixture of Experts

A long-cherished vision in robotics is to equip robots with skills that match the versatility and precision of humans. For example, when playing table tennis, a robot should be capable of returning the ball in various ways while precisely placing it at the desired location. A common approach to model such versatile behavior is to use a Mixture of Experts (MoE) model, where each expert is a contextual motion primitive. However, learning such MoEs is challenging as most objectives force the model to cover the entire context space, which prevents specialization of the primitives resulting in rather low-quality components. Starting from maximum entropy reinforcement learning (RL), we decompose the objective into optimizing an individual lower bound per mixture component. Further, we introduce a curriculum by allowing the components to focus on a local context region, enabling the model to learn highly accurate skill representations. To this end, we use local context distributions that are adapted jointly with the expert primitives. Our lower bound advocates an iterative addition of new components, where new components will concentrate on local context regions not covered by the current MoE. This local and incremental learning results in a modular MoE model of high accuracy and versatility, where both properties can be scaled by adding more components on the fly. We demonstrate this by an extensive ablation and on two challenging simulated robot skill learning tasks. We compare our achieved performance to LaDiPS and HiREPS, a known hierarchical policy search method for learning diverse skills.

cs.LG↗

arXiv2021

Human-machine Symbiosis: A Multivariate Perspective for Physically Coupled Human-machine Systems

The notion of symbiosis has been increasingly mentioned in research on physically coupled human-machine systems. Yet, a uniform specification on which aspects constitute human-machine symbiosis is missing. By combining the expertise of different disciplines, we elaborate on a multivariate perspective of symbiosis as the highest form of physically coupled human-machine systems. Four dimensions are considered: Task, interaction, performance, and experience. First, human and machine work together to accomplish a common task conceptualized on both a decision and an action level (task dimension). Second, each partner possesses an internal representation of own as well as the other partner's intentions and influence on the environment. This alignment, which is the core of the interaction, constitutes the symbiotic understanding between both partners, being the basis of a joint, highly coordinated and effective action (interaction dimension). Third, the symbiotic interaction leads to synergetic effects regarding the intention recognition and complementary strengths of the partners, resulting in a higher overall performance (performance dimension). Fourth, symbiotic systems specifically change the user's experiences, like flow, acceptance, sense of agency, and embodiment (experience dimension). This multivariate perspective is flexible and generic and is also applicable in diverse human-machine scenarios, helping to bridge barriers between different disciplines.

cs.HC↗

arXiv2021

Switching Recurrent Kalman Networks

Forecasting driving behavior or other sensor measurements is an essential component of autonomous driving systems. Often real-world multivariate time series data is hard to model because the underlying dynamics are nonlinear and the observations are noisy. In addition, driving data can often be multimodal in distribution, meaning that there are distinct predictions that are likely, but averaging can hurt model performance. To address this, we propose the Switching Recurrent Kalman Network (SRKN) for efficient inference and prediction on nonlinear and multi-modal time-series data. The model switches among several Kalman filters that model different aspects of the dynamics in a factorized latent state. We empirically test the resulting scalable and interpretable deep state-space model on toy data sets and real driving data from taxis in Porto. In all cases, the model can capture the multimodal nature of the dynamics in the data.

cs.LG↗

arXiv2021

Versatile Inverse Reinforcement Learning via Cumulative Rewards

Inverse Reinforcement Learning infers a reward function from expert demonstrations, aiming to encode the behavior and intentions of the expert. Current approaches usually do this with generative and uni-modal models, meaning that they encode a single behavior. In the common setting, where there are various solutions to a problem and the experts show versatile behavior this severely limits the generalization capabilities of these methods. We propose a novel method for Inverse Reinforcement Learning that overcomes these problems by formulating the recovered reward as a sum of iteratively trained discriminators. We show on simulated tasks that our approach is able to recover general, high-quality reward functions and produces policies of the same quality as behavioral cloning approaches designed for versatile behavior.

cs.LG↗

arXiv2021

Cooperative Assistance in Robotic Surgery through Multi-Agent Reinforcement Learning

Cognitive cooperative assistance in robot-assisted surgery holds the potential to increase quality of care in minimally invasive interventions. Automation of surgical tasks promises to reduce the mental exertion and fatigue of surgeons. In this work, multi-agent reinforcement learning is demonstrated to be robust to the distribution shift introduced by pairing a learned policy with a human team member. Multi-agent policies are trained directly from images in simulation to control multiple instruments in a sub task of the minimally invasive removal of the gallbladder. These agents are evaluated individually and in cooperation with humans to demonstrate their suitability as autonomous assistants. Compared to human teams, the hybrid teams with artificial agents perform better considering completion time (44.4% to 71.2% shorter) as well as number of collisions (44.7% to 98.0% fewer). Path lengths, however, increase under control of an artificial agent (11.4% to 33.5% longer). A multi-agent formulation of the learning problem was favored over a single-agent formulation on this surgical sub task, due to the sequential learning of the two instruments. This approach may be extended to other tasks that are difficult to formulate within the standard reinforcement learning framework. Multi-agent reinforcement learning may shift the paradigm of cognitive robotic surgery towards seamless cooperation between surgeons and assistive technologies.

cs.RO↗

arXiv2021

Residual Feedback Learning for Contact-Rich Manipulation Tasks with Uncertainty

While classic control theory offers state of the art solutions in many problem scenarios, it is often desired to improve beyond the structure of such solutions and surpass their limitations. To this end, residual policy learning (RPL) offers a formulation to improve existing controllers with reinforcement learning (RL) by learning an additive "residual" to the output of a given controller. However, the applicability of such an approach highly depends on the structure of the controller. Often, internal feedback signals of the controller limit an RL algorithm to adequately change the policy and, hence, learn the task. We propose a new formulation that addresses these limitations by also modifying the feedback signals to the controller with an RL policy and show superior performance of our approach on a contact-rich peg-insertion task under position and orientation uncertainty. In addition, we use a recent Cartesian impedance control architecture as the control framework which can be available to us as a black-box while assuming no knowledge about its input/output structure, and show the difficulties of standard RPL. Furthermore, we introduce an adaptive curriculum for the given task to gradually increase the task difficulty in terms of position and orientation uncertainty. A video showing the results can be found at https://youtu.be/SAZm_Krze7U .

cs.RO↗

arXiv2021

A Study on Dense and Sparse (Visual) Rewards in Robot Policy Learning

Deep Reinforcement Learning (DRL) is a promising approach for teaching robots new behaviour. However, one of its main limitations is the need for carefully hand-coded reward signals by an expert. We argue that it is crucial to automate the reward learning process so that new skills can be taught to robots by their users. To address such automation, we consider task success classifiers using visual observations to estimate the rewards in terms of task success. In this work, we study the performance of multiple state-of-the-art deep reinforcement learning algorithms under different types of reward: Dense, Sparse, Visual Dense, and Visual Sparse rewards. Our experiments in various simulation tasks (Pendulum, Reacher, Pusher, and Fetch Reach) show that while DRL agents can learn successful behaviours using visual rewards when the goal targets are distinguishable, their performance may decrease if the task goal is not clearly visible. Our results also show that visual dense rewards are more successful than visual sparse rewards and that there is no single best algorithm for all tasks.

cs.RO↗

arXiv2021

Navigate-and-Seek: a Robotics Framework for People Localization in Agricultural Environments

The agricultural domain offers a working environment where many human laborers are nowadays employed to maintain or harvest crops, with huge potential for productivity gains through the introduction of robotic automation. Detecting and localizing humans reliably and accurately in such an environment, however, is a prerequisite to many services offered by fleets of mobile robots collaborating with human workers. Consequently, in this paper, we expand on the concept of a topological particle filter (TPF) to accurately and individually localize and track workers in a farm environment, integrating information from heterogeneous sensors and combining local active sensing (exploiting a robot's onboard sensing employing a Next-Best-Sense planning approach) and global localization (using affordable IoT GNSS devices). We validate the proposed approach in topologies created for the deployment of robotics fleets to support fruit pickers in a real farm environment. By combining multi-sensor observations on the topological level complemented by active perception through the NBS approach, we show that we can improve the accuracy of picker localization in comparison to prior work.

cs.RO↗

arXiv2021

Learning Riemannian Manifolds for Geodesic Motion Skills

For robots to work alongside humans and perform in unstructured environments, they must learn new motion skills and adapt them to unseen situations on the fly. This demands learning models that capture relevant motion patterns, while offering enough flexibility to adapt the encoded skills to new requirements, such as dynamic obstacle avoidance. We introduce a Riemannian manifold perspective on this problem, and propose to learn a Riemannian manifold from human demonstrations on which geodesics are natural motion skills. We realize this with a variational autoencoder (VAE) over the space of position and orientations of the robot end-effector. Geodesic motion skills let a robot plan movements from and to arbitrary points on the data manifold. They also provide a straightforward method to avoid obstacles by redefining the ambient metric in an online fashion. Moreover, geodesics naturally exploit the manifold resulting from multiple--mode tasks to design motions that were not explicitly demonstrated previously. We test our learning framework using a 7-DoF robotic manipulator, where the robot satisfactorily learns and reproduces realistic skills featuring elaborated motion patterns, avoids previously unseen obstacles, and generates novel movements in multiple-mode settings.

cs.RO↗

arXiv2021

Differentiable Robust LQR Layers

This paper proposes a differentiable robust LQR layer for reinforcement learning and imitation learning under model uncertainty and stochastic dynamics. The robust LQR layer can exploit the advantages of robust optimal control and model-free learning. It provides a new type of inductive bias for stochasticity and uncertainty modeling in control systems. In particular, we propose an efficient way to differentiate through a robust LQR optimization program by rewriting it as a convex program (i.e. semi-definite program) of the worst-case cost. Based on recent work on using convex optimization inside neural network layers, we develop a fully differentiable layer for optimizing this worst-case cost, i.e. we compute the derivative of a performance measure w.r.t the model's unknown parameters, model uncertainty and stochasticity parameters. We demonstrate the proposed method on imitation learning and approximate dynamic programming on stochastic and uncertain domains. The experiment results show that the proposed method can optimize robust policies under uncertain situations, and are able to achieve a significantly better performance than existing methods that do not model uncertainty directly.

cs.RO↗

arXiv2021

Probabilistic approach to physical object disentangling

Physically disentangling entangled objects from each other is a problem encountered in waste segregation or in any task that requires disassembly of structures. Often there are no object models, and, especially with cluttered irregularly shaped objects, the robot can not create a model of the scene due to occlusion. One of our key insights is that based on previous sensory input we are only interested in moving an object out of the disentanglement around obstacles. That is, we only need to know where the robot can successfully move in order to plan the disentangling. Due to the uncertainty we integrate information about blocked movements into a probability map. The map defines the probability of the robot successfully moving to a specific configuration. Using as cost the failure probability of a sequence of movements we can then plan and execute disentangling iteratively. Since our approach circumvents only previously encountered obstacles, new movements will yield information about unknown obstacles that block movement until the robot has learned to circumvent all obstacles and disentangling succeeds. In the experiments, we use a special probabilistic version of the Rapidly exploring Random Tree (RRT) algorithm for planning and demonstrate successful disentanglement of objects both in 2-D and 3-D simulation, and, on a KUKA LBR 7-DOF robot. Moreover, our approach outperforms baseline methods.

cs.RO↗