Toyota Motor Corporation (Belgium)

The development of driver monitoring systems (DMS) has emerged as a pivotal advancement in automotive safety to monitor driver behaviors and prevent road accidents. Driver emotions are one of the critical factors that affect driving safety. While numerous studies have explored facial macro-expressions (FMaEs) of drivers, these facial expressions can be consciously controlled, allowing individuals to conceal their true negative emotions. However, inaccurate capturing of true emotions can lead to severe accidents. In this paper, we focus on facial micro-expressions (FMEs), which are brief and involuntary facial movements that reveal deeper insights into genuine emotions. Thereby, they provide a more accurate measure of a driver's emotional condition, and are crucial for assessing driver readiness and safety. To facilitate FME studies of drivers, we analyzed driver videos from an affective driver-pedestrian interaction experiment, providing corresponding facial action units and FME annotations, and built a real driver FME dataset. By analyzing the driver data, we found that there are more FMEs when drivers respond to negative stimuli compared with positive stimuli, and there are more FMaEs when drivers respond to positive stimuli compared with negative stimuli. To protect driver identities while maintaining the integrity of the facial movements, we released a synthesized driver FME dataset. Furthermore, we benchmarked deep learning-based architectures on the real and synthetic driver FME datasets, uncovering the challenges inherent in recognizing FMEs in real-world driving scenarios.

The comprehensive characterization of the structure-electrocatalytic activity relationships is of great interest to support the development of efficient catalysts for hydrogen evolution reaction (HER) in proton-exchange membrane (PEM) electrolyzers. The successful implementation of a bimetallic iron-molybdenum sulfide electrocatalyst for HER into a PEM electrolyzer has been reported. However, the precise nature of their active sites is still elusive. In this study, we employed X-ray absorption spectroscopy (XAS) to study iron-molybdenum and molybdenum sulfides obtained by a microwave irradiation synthetic approach. Operando XAS measurements in an acidic environment revealed sulfur-defective amorphous MoSx and FeS phases. Although the amount of dopant Fe is 10 times lower than that of Mo, the synergy between the two phases promotes the HER performance of iron-molybdenum sulfides.

High-throughput compositional screening of Cs3M2X9 compounds based on anion exchanges yielded one hundred single-phase products with independently tunable compositions of M = Bi, Sb, and Bi + Sb and X = Cl, Cl + Br, Br, Br + I, I, and Cl + Br + I.

To fully leverage the capabilities of mobile manipulation robots, it is imperative that they are able to autonomously execute long-horizon tasks in large unexplored environments. While large language models (LLMs) have shown emergent reasoning skills on arbitrary tasks, existing work primarily concentrates on explored environments, typically focusing on either navigation or manipulation tasks in isolation. In this work, we propose MoMa-LLM, a novel approach that grounds language models within structured representations derived from open-vocabulary scene graphs, dynamically updated as the environment is explored. We tightly interleave these representations with an object-centric action space. Given object detections, the resulting approach is zero-shot, open-vocabulary, and readily extendable to a spectrum of mobile manipulation and household robotic tasks. We demonstrate the effectiveness of MoMa-LLM in a novel semantic interactive search task in large realistic indoor environments. In extensive experiments in both simulation and the real world, we show substantially improved search efficiency compared to conventional baselines and state-of-the-art approaches, as well as its applicability to more abstract tasks.

A coil-based grey box model is developed for high-frequency modeling of permanent magnet synchronous machines (PMSMs) with interior magnets in rotors to predict the voltage distribution along stator winding. Unlike the grey box model in the literature, the developed coil-based model can predict the voltage distribution along stator winding. Compared to the white box model, no complex modeling based on finite element simulation results is required. As the name suggests, the developed model has each winding coil as a basis, and the entire model is created by connecting coils according to the winding schema. Furthermore, the coil-based model considers frequency-dependent inductive and capacitive coupling within electrical machines, including skin and proximity effects. The model is parametrized using an optimization program based on measured impedance spectra at machine terminals. As a result, an average deviation of less than 4 % is observed in the impedance modeling. With the coil-based model's help, the common mode voltage distribution is predicted with almost zero errors between modeled and measured voltage peak values. Regarding the differential mode voltage distribution, the voltage maximum can be predicted with a maximum relative error of less than 10 % in the time domain.

Conventional distributed approaches to coverage control may suffer from lack of convergence and poor performance, due to the fact that agents have limited information, especially in non-convex discrete environments. To address this issue, we extend the approach of [12] which demonstrates how a limited degree of inter-agent communication can be exploited to overcome such pitfalls in one-dimensional discrete environments. The focus of this paper is on extending such results to general dimensional settings. We show that the extension is convergent and keeps the approximation ratio of 2, meaning that any stable solution is guaranteed to have a performance within 50% of the optimal one. The experimental results exhibit that our algorithm outperforms several state-of-the-art algorithms, and also that the runtime is scalable.

ABSTRACT Existing mathematical driver models with steering torque feedback are reviewed. It is argued that a new linear parametric driver-steering-vehicle model is required to provide greater insight to subjective assessment of steering torque feedback. The proposed new model incorporates: neuromuscular dynamics; proprioceptive (muscle stretch) and visual sensory dynamics; steering torque perception via efference copy; process and measurement noise; state estimator; and an internal predictive model of the plant. A series of path-following experiments with thirteen test subjects is performed on a fixed-base driving simulator. A Box-Jenkins model accounted for 90% of the variance of the measured handwheel angle time series averaged over the test subjects (VAF). The unknown parameter values of the model were identified from the driving simulator data. The prediction accuracy approached the upper bound given by the Box-Jenkins model. Process noise was found to be linearly dependent on muscle activation torque, and this signal-dependent behaviour was also present in the measurement noise. The model was re-identified using signal to noise ratios as parameters. The VAF was 86.6%, close to that of the Box-Jenkins model, and the identified parameter values are physically plausible, giving confidence that the model is suitable for further work on understanding steering torque feedback.

Current self-supervised approaches for skeleton action representation learning often focus on constrained scenarios, where videos and skeleton data are recorded in laboratory settings. When dealing with estimated skeleton data in realworld videos, such methods perform poorly due to the large variations across subjects and camera viewpoints. To address this issue, we introduce ViA, a novel View-Invariant Autoencoder for self-supervised skeleton action representation learning 1 . ViA leverages motion retargeting between different human performers as a pretext task, in order to disentangle the latent action-specific 'Motion' features on top of the visual representation of a 2D or 3D skeleton sequence. Such 'Motion' features are invariant to skeleton geometry and camera view and allow ViA to facilitate both, cross-subject and cross-view action classification tasks. We conduct a study focusing on transfer-learning for skeleton-based action recognition with self-supervised pre-training on real-world data (e.g., Posetics). Our results showcase that skeleton representations learned from ViA are generic enough to improve upon stateof-the-art action classification accuracy, not only on 3D laboratory datasets such as NTU-RGB+D 60 and NTU-RGB+D 120, but also on real-world datasets where only 2D data are accurately estimated, e.g., Toyota Smarthome, UAV-Human and Penn Action.

The alloying of two-dimensional (2D) transition metal dichalcogenides (TMDs) is an established route to produce robust semiconductors with continuously tunable optoelectronic properties. However, typically reported methods for fabricating alloyed 2D TMD nanosheets are not suitable for the inexpensive, scalable production of large-area (m2) devices. Herein we describe a general method to afford large quantities of compositionally-tunable 2D TMD nanosheets using commercially available powders and liquid-phase exfoliation. Beginning with Mo(1-x)WxS2 nanosheets, we demonstrate tunable optoelectronic properties as a function of composition. We extend this method to produce Mo0.5W0.5Se2 MoSSe, WSSe, and quaternary Mo0.5W0.5SSe nanosheets. High-resolution scanning transmission electron microscopy (STEM) imaging confirms the atomic arrangement of the nanosheets, while an array of spectroscopic techniques is used to characterize the chemical and optoelectronic properties. This transversal method represents an important step towards upscaling tailored TMD nanosheets with a broad range of tunable optoelectronic properties for large-area devices.