Search

Egocentric vision aims to capture and analyse the world from the first-person perspective. We explore the possibilities for egocentric wearable devices to improve and enhance industrial use cases w.r.t. data collection, annotation, labelling and downstream applications. This would contribute to easier data collection and allow users to provide additional context. We envision that this approach could serve as a supplement to the traditional industrial Machine Vision workflow. Code, Dataset and related resources will be available at: https://github.com/Vivek9Chavan/EgoVis24, Comment: To be presented at the First Joint Egocentric Vision (EgoVis) Workshop, held in conjunction with CVPR 2024

Surface nanostructures provide the possibility to create and tailor surface functionalities mainly via controlling their topography along with other chemical and physical material properties. One of the most appealing technologies for surface functionalization via micro- and nanostructuring is based on laser processing. This can be done either via direct contour-shaping of the irradiated material using a tightly focused laser beam or in a self-ordered way that allows employing larger laser beam diameters along with areal scanning to create a variety of laser-induced periodic surface structures (LIPSS). For the latter approach, particularly ultrashort pulsed lasers have recently pushed the borders across long-lasting limitations regarding the minimum achievable feature sizes and additionally boosted up the production times. This chapter reviews the plethora of recently investigated applications of LIPSS-for example, via imposing diffractive or plasmonic structural colors, the management of liquids and surface wetting properties, biomedical and bioinspired functionalities, beneficial effects in tribology for reducing friction and wear, the manipulation of optical scattering and absorption in photovoltaics, or the modification of magnetic or superconducting surface properties in other energy applications. The footprint of the LIPSS-based technology is explored in detail regarding the current state of industrialization, including an analysis of the market and associated LIPSS production costs.

The operation of industrial facilities is a broad field for optimization. Industrial plants are often a) composed of several components, b) linked using network technology, c) physically interconnected and d) complex regarding the effect of set-points and operating points in every entity. This leads to the possibility of overall optimization but also to a high complexity of the emerging optimization problems. The decomposition of complex systems allows the modeling of individual models which can be structured according to the physical topology. A method for energy performance indicators (EnPI) helps to formulate an optimization problem. The optimization algorithm OptTopo achieves efficient set-points by traversing a graph representation of the overall system.

Force control enables hands-on teaching and physical collaboration, with the potential to improve ergonomics and flexibility of automation. Established methods for the design of compliance, impedance control, and \rev{collision response} can achieve free-space stability and acceptable peak contact force on lightweight, lower payload robots. Scaling collaboration to higher payloads can allow new applications, but introduces challenges due to the more significant payload dynamics and the use of higher-payload industrial robots. To achieve high-payload manual guidance with contact, this paper proposes and validates new mechatronic design methods: standard admittance control is extended with damping feedback, compliant structures are integrated to the environment, and a contact response method which allows continuous admittance control is proposed. These methods are compared with respect to free-space stability, contact stability, and peak contact force. The resulting methods are then applied to realize two contact-rich tasks on a 16 kg payload (peg in hole and slot assembly) and free-space co-manipulation of a 50 kg payload., Comment: Accepted at IEEE RA-L

Identifying changes in contact during contact-rich manipulation can detect task state or errors, enabling improved robustness and autonomy. The ability to detect contact is affected by the mechatronic design of the robot, especially its physical compliance. Established methods can design physical compliance for many aspects of contact performance (e.g. peak contact force, motion/force control bandwidth), but are based on time-invariant dynamic models. A change in contact mode is a discrete change in coupled robot-environment dynamics, not easily considered in existing design methods. Towards designing robots which can robustly detect changes in contact mode online, this paper investigates how mechatronic design can improve contact estimation, with a focus on the impact of the location and degree of compliance. A design metric of information gain is proposed which measures how much position/force measurements reduce uncertainty in the contact mode estimate. This information gain is developed for fully- and partially-observed systems, as partial observability can arise from joint flexibility in the robot or environmental inertia. Hardware experiments with various compliant setups validate that information gain predicts the speed and certainty with which contact is detected in (i) monitoring of contact-rich assembly and (ii) collision detection., Comment: On review, video: https://youtu.be/cPLsjyHVrh8

Learned representations of dynamical systems reduce dimensionality, potentially supporting downstream reinforcement learning (RL). However, no established methods predict a representation's suitability for control and evaluation is largely done via downstream RL performance, slowing representation design. Towards a principled evaluation of representations for control, we consider the relationship between the true state and the corresponding representations, proposing that ideally each representation corresponds to a unique true state. This motivates two metrics: temporal smoothness and high mutual information between true state/representation. These metrics are related to established representation objectives, and studied on Lagrangian systems where true state, information requirements, and statistical properties of the state can be formalized for a broad class of systems. These metrics are shown to predict reinforcement learning performance in a simulated peg-in-hole task when comparing variants of autoencoder-based representations., Comment: 10 pages, 8 figures, associated video at https://youtu.be/adHXi58u9qw. Submitted to ICRA 2021. v2 24.Nov.2020: Typographic corrections

Uncertainty quantification is an important and challenging problem in deep learning. Previous methods rely on dropout layers which are not present in modern deep architectures or batch normalization which is sensitive to batch sizes. In this work, we address the problem of uncertainty quantification in deep residual networks by using a regularization technique called stochastic depth. We show that training residual networks using stochastic depth can be interpreted as a variational approximation to the intractable posterior over the weights in Bayesian neural networks. We demonstrate that by sampling from a distribution of residual networks with varying depth and shared weights, meaningful uncertainty estimates can be obtained. Moreover, compared to the original formulation of residual networks, our method produces well-calibrated softmax probabilities with only minor changes to the network's structure. We evaluate our approach on popular computer vision datasets and measure the quality of uncertainty estimates. We also test the robustness to domain shift and show that our method is able to express higher predictive uncertainty on out-of-distribution samples. Finally, we demonstrate how the proposed approach could be used to obtain uncertainty estimates in facial verification applications.

Hysteresis-controlled devices are widely used in industrial applications. For example, cooling devices usually contain a two-point controller, resulting in a nonlinear hybrid system with two discrete states. Dynamic models of systems are essential for optimizing such industrial supply technology. However, conventional system identification approaches can hardly handle hysteresis-controlled devices. Thus, the new identification method Sparse Identification of Nonlinear Dynamics (SINDy) is extended to consider hybrid systems. SINDy composes models from basis functions out of a customized library in a data-driven manner. For modeling systems that behave dependent on their own past as in the case of natural hysteresis, Ferenc Preisach introduced the relay hysteron as an elementary mathematical description. In this new method (SINDyHybrid), tailored basis functions in form of relay hysterons are added to the library which is used by SINDy. Experiments with a hysteresis controlled water basin show that this approach correctly identifies state transitions of hybrid systems and also succeeds in modeling the dynamics of the discrete system states. A novel proximity hysteron achieves the robustness of this method. The impacts of the sampling rate and the signal noise ratio of the measurement data are examined accordingly.

The possibility to excite Surface Plasmon Polaritons (SPPs) at the interface between two media depends on the optical properties of both media and geometrical aspects. Specific conditions allowing the coupling of light with a plasmon-active interface must be satisfied. Plasmonic effects are well described in noble metals where the imaginary part of the dielectric permittivity is often neglected ("perfect medium approximation"). However, some systems exist for which such approximation cannot be applied, hence requiring a refinement of the common SPP theory. In this context, several properties of SPPs such as excitation conditions, period of the electromagnetic field modulation and SPP lifetime then may strongly deviate from that of the perfect medium approximation. In this paper, calculations taking into account the imaginary part of the dielectric permittivities are presented. The model identifies analytical terms which should not be neglected in the mathematical description of SPPs on lossy materials. These calculations are applied to numerous material combinations resulting in a prediction of the corresponding SPP features. A list of plasmon-active interfaces is provided along with a quantification of the above mentioned SPP properties in the regime where the perfect medium approximation is not applicable., Comment: 9 pages, 7 figures, regular paper