Teruhisa Misu

Lingfeng Sun, Chen Tang, Yaru Niu et al. · berkeley

Motion forecasting in highly interactive scenarios is a challenging problem in autonomous driving. In such scenarios, we need to accurately predict the joint behavior of interacting agents to ensure the safe and efficient navigation of autonomous vehicles. Recently, goal-conditioned methods have gained increasing attention due to their advantage in performance and their ability to capture the multimodality in trajectory distribution. In this work, we study the joint trajectory prediction problem with the goal-conditioned framework. In particular, we introduce a conditional-variational-autoencoder-based (CVAE) model to explicitly encode different interaction modes into the latent space. However, we discover that the vanilla model suffers from posterior collapse and cannot induce an informative latent space as desired. To address these issues, we propose a novel approach to avoid KL vanishing and induce an interpretable interactive latent space with pseudo labels. The proposed pseudo labels allow us to incorporate domain knowledge on interaction in a flexible manner. We motivate the proposed method using an illustrative toy example. In addition, we validate our framework on the Waymo Open Motion Dataset with both quantitative and qualitative evaluations.

Yuning Qiu, Teruhisa Misu, Carlos Busso

Anomaly driving detection is an important problem in advanced driver assistance systems (ADAS). It is important to identify potential hazard scenarios as early as possible to avoid potential accidents. This study proposes an unsupervised method to quantify driving anomalies using a conditional generative adversarial network (GAN). The approach predicts upcoming driving scenarios by conditioning the models on the previously observed signals. The system uses the difference of the output from the discriminator between the predicted and actual signals as a metric to quantify the anomaly degree of a driving segment. We take a driver-centric approach, considering physiological signals from the driver and controller area network-Bus (CAN-Bus) signals from the vehicle. The approach is implemented with convolutional neural networks (CNNs) to extract discriminative feature representations, and with long short-term memory (LSTM) cells to capture temporal information. The study is implemented and evaluated with the driving anomaly dataset (DAD), which includes 250 hours of naturalistic recordings manually annotated with driving events. The experimental results reveal that recordings annotated with events that are likely to be anomalous, such as avoiding on-road pedestrians and traffic rule violations, have higher anomaly scores than recordings without any event annotation. The results are validated with perceptual evaluations, where annotators are asked to assess the risk and familiarity of the videos detected with high anomaly scores. The results indicate that the driving segments with higher anomaly scores are more risky and less regularly seen on the road than other driving segments, validating the proposed unsupervised approach.

Kaiwen Zhou, Kwonjoon Lee, Teruhisa Misu et al.

In our work, we explore the synergistic capabilities of pre-trained vision-and-language models (VLMs) and large language models (LLMs) on visual commonsense reasoning (VCR) problems. We find that VLMs and LLMs-based decision pipelines are good at different kinds of VCR problems. Pre-trained VLMs exhibit strong performance for problems involving understanding the literal visual content, which we noted as visual commonsense understanding (VCU). For problems where the goal is to infer conclusions beyond image content, which we noted as visual commonsense inference (VCI), VLMs face difficulties, while LLMs, given sufficient visual evidence, can use commonsense to infer the answer well. We empirically validate this by letting LLMs classify VCR problems into these two categories and show the significant difference between VLM and LLM with image caption decision pipelines on two subproblems. Moreover, we identify a challenge with VLMs' passive perception, which may miss crucial context information, leading to incorrect reasoning by LLMs. Based on these, we suggest a collaborative approach, named ViCor, where pre-trained LLMs serve as problem classifiers to analyze the problem category, then either use VLMs to answer the question directly or actively instruct VLMs to concentrate on and gather relevant visual elements to support potential commonsense inferences. We evaluate our framework on two VCR benchmark datasets and outperform all other methods that do not require in-domain fine-tuning.

Zhaobo K. Zheng, Kumar Akash, Teruhisa Misu et al.

A key factor to optimal acceptance and comfort of automated vehicle features is the driving style. Mismatches between the automated and the driver preferred driving styles can make users take over more frequently or even disable the automation features. This work proposes identification of user driving style preference with multimodal signals, so the vehicle could match user preference in a continuous and automatic way. We conducted a driving simulator study with 36 participants and collected extensive multimodal data including behavioral, physiological, and situational data. This includes eye gaze, steering grip force, driving maneuvers, brake and throttle pedal inputs as well as foot distance from pedals, pupil diameter, galvanic skin response, heart rate, and situational drive context. Then, we built machine learning models to identify preferred driving styles, and confirmed that all modalities are important for the identification of user preference. This work paves the road for implicit adaptive driving styles on automated vehicles.

Xiaofeng Gao, Xingwei Wu, Samson Ho et al.

Although partially autonomous driving (AD) systems are already available in production vehicles, drivers are still required to maintain a sufficient level of situational awareness (SA) during driving. Previous studies have shown that providing information about the AD's capability using user interfaces can improve the driver's SA. However, displaying too much information increases the driver's workload and can distract or overwhelm the driver. Therefore, to design an efficient user interface (UI), it is necessary to understand its effect under different circumstances. In this paper, we focus on a UI based on augmented reality (AR), which can highlight potential hazards on the road. To understand the effect of highlighting on drivers' SA for objects with different types and locations under various traffic densities, we conducted an in-person experiment with 20 participants on a driving simulator. Our study results show that the effects of highlighting on drivers' SA varied by traffic densities, object locations and object types. We believe our study can provide guidance in selecting which object to highlight for the AR-based driver-assistance interface to optimize SA for drivers driving and monitoring partially autonomous vehicles.

Kaihong Wang, Kumar Akash, Teruhisa Misu

Unpaired video-to-video translation aims to translate videos between a source and a target domain without the need of paired training data, making it more feasible for real applications. Unfortunately, the translated videos generally suffer from temporal and semantic inconsistency. To address this, many existing works adopt spatiotemporal consistency constraints incorporating temporal information based on motion estimation. However, the inaccuracies in the estimation of motion deteriorate the quality of the guidance towards spatiotemporal consistency, which leads to unstable translation. In this work, we propose a novel paradigm that regularizes the spatiotemporal consistency by synthesizing motions in input videos with the generated optical flow instead of estimating them. Therefore, the synthetic motion can be applied in the regularization paradigm to keep motions consistent across domains without the risk of errors in motion estimation. Thereafter, we utilize our unsupervised recycle and unsupervised spatial loss, guided by the pseudo-supervision provided by the synthetic optical flow, to accurately enforce spatiotemporal consistency in both domains. Experiments show that our method is versatile in various scenarios and achieves state-of-the-art performance in generating temporally and semantically consistent videos. Code is available at: https://github.com/wangkaihong/Unsup_Recycle_GAN/.

Ahmed Taha, Yi-Ting Chen, Teruhisa Misu et al.

We employ triplet loss as a feature embedding regularizer to boost classification performance. Standard architectures, like ResNet and Inception, are extended to support both losses with minimal hyper-parameter tuning. This promotes generality while fine-tuning pretrained networks. Triplet loss is a powerful surrogate for recently proposed embedding regularizers. Yet, it is avoided due to large batch-size requirement and high computational cost. Through our experiments, we re-assess these assumptions. During inference, our network supports both classification and embedding tasks without any computational overhead. Quantitative evaluation highlights a steady improvement on five fine-grained recognition datasets. Further evaluation on an imbalanced video dataset achieves significant improvement. Triplet loss brings feature embedding characteristics like nearest neighbor to classification models. Code available at \url{http://bit.ly/2LNYEqL}.

Benjamin Panny, Shashank Mehrotra, Zahra Zahedi et al.

Computational models of collaboration without prior coordination often overlook how heterogeneous agent traits and complex task structures jointly produce systemic bottlenecks, inefficiencies, and contribution inequalities. We address this by using an agent-based model of ad-hoc teamwork in a kitchen environment. Our model integrates diverse agent personas with tasks that combine serial and parallel dependencies. We identify a specialist's dilemma, where rigid role assertion generates system-level bottlenecks, amplifies workload inequality, and fosters fragmented, homophilous networks. We also find that team size and communication overhead interact with problem structure to generate diminishing returns and redundant collaboration. Linking micro-level behavior to macro-level outcomes provides insights into emergent collaboration and design principles for effective multi-agent teamwork.

Chenran Li, Aolin Xu, Enna Sachdeva et al.

The driver warning system that alerts the human driver about potential risks during driving is a key feature of an advanced driver assistance system. Existing driver warning technologies, mainly the forward collision warning and unsafe lane change warning, can reduce the risk of collision caused by human errors. However, the current design methods have several major limitations. Firstly, the warnings are mainly generated in a one-shot manner without modeling the ego driver's reactions and surrounding objects, which reduces the flexibility and generality of the system over different scenarios. Additionally, the triggering conditions of warning are mostly rule-based threshold-checking given the current state, which lacks the prediction of the potential risk in a sufficiently long future horizon. In this work, we study the problem of optimally generating driver warnings by considering the interactions among the generated warning, the driver behavior, and the states of ego and surrounding vehicles on a long horizon. The warning generation problem is formulated as a partially observed Markov decision process (POMDP). An optimal warning generation framework is proposed as a solution to the proposed POMDP. The simulation experiments demonstrate the superiority of the proposed solution to the existing warning generation methods.

Minxue Niu, Zhaobo Zheng, Kumar Akash et al.

Humans' internal states play a key role in human-machine interaction, leading to the rise of human state estimation as a prominent field. Compared to swift state changes such as surprise and irritation, modeling gradual states like trust and satisfaction are further challenged by label sparsity: long time-series signals are usually associated with a single label, making it difficult to identify the critical span of state shifts. Windowing has been one widely-used technique to enable localized analysis of long time-series data. However, the performance of downstream models can be sensitive to the window size, and determining the optimal window size demands domain expertise and extensive search. To address this challenge, we propose a Selective Windowing Attention Network (SWAN), which employs window prompts and masked attention transformation to enable the selection of attended intervals with flexible lengths. We evaluate SWAN on the task of trust prediction on a new multimodal driving simulation dataset. Experiments show that SWAN significantly outperforms an existing empirical window selection baseline and neural network baselines including CNN-LSTM and Transformer. Furthermore, it shows robustness across a wide span of windowing ranges, compared to the traditional windowing approach.

Abinay Reddy Naini, Zhaobo K. Zheng, Teruhisa Misu et al.

Human state detection and behavior prediction have seen significant advancements with the rise of machine learning and multimodal sensing technologies. However, predicting prosocial behavior intentions in mobility scenarios, such as helping others on the road, is an underexplored area. Current research faces a major limitation. There are no large, labeled datasets available for prosocial behavior, and small-scale datasets make it difficult to train deep-learning models effectively. To overcome this, we propose a self-supervised learning approach that harnesses multi-modal data from existing physiological and behavioral datasets. By pre-training our model on diverse tasks and fine-tuning it with a smaller, manually labeled prosocial behavior dataset, we significantly enhance its performance. This method addresses the data scarcity issue, providing a more effective benchmark for prosocial behavior prediction, and offering valuable insights for improving intelligent vehicle systems and human-machine interaction.

Kei Katsumata, Yui Iioka, Naoki Hosomi et al.

We focus on the task of identifying the location of target regions from a natural language instruction and a front camera image captured by a mobility. This task is challenging because it requires both existence prediction and segmentation, particularly for stuff-type target regions with ambiguous boundaries. Existing methods often underperform in handling stuff-type target regions, in addition to absent or multiple targets. To overcome these limitations, we propose GENNAV, which predicts target existence and generates segmentation masks for multiple stuff-type target regions. To evaluate GENNAV, we constructed a novel benchmark called GRiN-Drive, which includes three distinct types of samples: no-target, single-target, and multi-target. GENNAV achieved superior performance over baseline methods on standard evaluation metrics. Furthermore, we conducted real-world experiments with four automobiles operated in five geographically distinct urban areas to validate its zero-shot transfer performance. In these experiments, GENNAV outperformed baseline methods and demonstrated its robustness across diverse real-world environments. The project page is available at https://gennav.vercel.app/.

Zahra Zahedi, Shashank Mehrotra, Teruhisa Misu et al.

For future human-autonomous vehicle (AV) interactions to be effective and smooth, human-aware systems that analyze and align human needs with automation decisions are essential. Achieving this requires systems that account for human cognitive states. We present a novel computational model in the form of a Dynamic Bayesian Network (DBN) that infers the cognitive states of both AV users and other road users, integrating this information into the AV's decision-making process. Specifically, our model captures the well-being of both an AV user and an interacting road user as cognitive states alongside trust. Our DBN models infer beliefs over the AV user's evolving well-being, trust, and intention states, as well as the possible well-being of other road users, based on observed interaction experiences. Using data collected from an interaction study, we refine the model parameters and empirically assess its performance. Finally, we extend our model into a causal inference model (CIM) framework for AV decision-making, enabling the AV to enhance user well-being and trust while balancing these factors with its own operational costs and the well-being of interacting road users. Our evaluation demonstrates the model's effectiveness in accurately predicting user's states and guiding informed, human-centered AV decisions.

Wenda Qin, Teruhisa Misu, Derry Wijaya

Vision-and-Language Navigation (VLN) is a challenging task in the field of artificial intelligence. Although massive progress has been made in this task over the past few years attributed to breakthroughs in deep vision and language models, it remains tough to build VLN models that can generalize as well as humans. In this paper, we provide a new perspective to improve VLN models. Based on our discovery that snapshots of the same VLN model behave significantly differently even when their success rates are relatively the same, we propose a snapshot-based ensemble solution that leverages predictions among multiple snapshots. Constructed on the snapshots of the existing state-of-the-art (SOTA) model $\circlearrowright$BERT and our past-action-aware modification, our proposed ensemble achieves the new SOTA performance in the R2R dataset challenge in Navigation Error (NE) and Success weighted by Path Length (SPL).

Haibei Zhu, Teruhisa Misu, Sujitha Martin et al.

Situation awareness (SA) is generally considered as the perception, understanding, and projection of objects' properties and positions. We believe if the system can sense drivers' SA, it can appropriately provide warnings for objects that drivers are not aware of. To investigate drivers' awareness, in this study, a human-subject experiment of driving simulation was conducted for data collection. While a previous predictive model for drivers' situation awareness utilized drivers' gaze movement only, this work utilizes object properties, characteristics of human visual sensory and memory mechanism. As a result, the proposed driver SA prediction model achieves over 70% accuracy and outperforms the baselines.

Jundi Liu, Kumar Akash, Teruhisa Misu et al.

Trust calibration is necessary to ensure appropriate user acceptance in advanced automation technologies. A significant challenge to achieve trust calibration is to quantitatively estimate human trust in real-time. Although multiple trust models exist, these models have limited predictive performance partly due to individual differences in trust dynamics. A personalized model for each person can address this issue, but it requires a significant amount of data for each user. We present a methodology to develop customized model by clustering humans based on their trust dynamics. The clustering-based method addresses the individual differences in trust dynamics while requiring significantly less data than personalized model. We show that our clustering-based customized models not only outperform the general model based on entire population, but also outperform simple demographic factor-based customized models. Specifically, we propose that two models based on ``confident'' and ``skeptical'' group of participants, respectively, can represent the trust behavior of the population. The ``confident'' participants, as compared to the ``skeptical'' participants, have higher initial trust levels, lose trust slower when they encounter low reliability operations, and have higher trust levels during trust-repair after the low reliability operations. In summary, clustering-based customized models improve trust prediction performance for further trust calibration considerations.

Kumar Akash, Neera Jain, Teruhisa Misu

Properly calibrated human trust is essential for successful interaction between humans and automation. However, while human trust calibration can be improved by increased automation transparency, too much transparency can overwhelm human workload. To address this tradeoff, we present a probabilistic framework using a partially observable Markov decision process (POMDP) for modeling the coupled trust-workload dynamics of human behavior in an action-automation context. We specifically consider hands-off Level 2 driving automation in a city environment involving multiple intersections where the human chooses whether or not to rely on the automation. We consider automation reliability, automation transparency, and scene complexity, along with human reliance and eye-gaze behavior, to model the dynamics of human trust and workload. We demonstrate that our model framework can appropriately vary automation transparency based on real-time human trust and workload belief estimates to achieve trust calibration.

Jinkyu Kim, Teruhisa Misu, Yi-Ting Chen et al.

Recent success suggests that deep neural control networks are likely to be a key component of self-driving vehicles. These networks are trained on large datasets to imitate human actions, but they lack semantic understanding of image contents. This makes them brittle and potentially unsafe in situations that do not match training data. Here, we propose to address this issue by augmenting training data with natural language advice from a human. Advice includes guidance about what to do and where to attend. We present the first step toward advice giving, where we train an end-to-end vehicle controller that accepts advice. The controller adapts the way it attends to the scene (visual attention) and the control (steering and speed). Attention mechanisms tie controller behavior to salient objects in the advice. We evaluate our model on a novel advisable driving dataset with manually annotated human-to-vehicle advice called Honda Research Institute-Advice Dataset (HAD). We show that taking advice improves the performance of the end-to-end network, while the network cues on a variety of visual features that are provided by advice. The dataset is available at https://usa.honda-ri.com/HAD.

Vidyasagar Sadhu, Teruhisa Misu, Dario Pompili

Corner cases are the main bottlenecks when applying Artificial Intelligence (AI) systems to safety-critical applications. An AI system should be intelligent enough to detect such situations so that system developers can prepare for subsequent planning. In this paper, we propose semi-supervised anomaly detection considering the imbalance of normal situations. In particular, driving data consists of multiple positive/normal situations (e.g., right turn, going straight), some of which (e.g., U-turn) could be as rare as anomalous situations. Existing machine learning based anomaly detection approaches do not fare sufficiently well when applied to such imbalanced data. In this paper, we present a novel multi-task learning based approach that leverages domain-knowledge (maneuver labels) for anomaly detection in driving data. We evaluate the proposed approach both quantitatively and qualitatively on 150 hours of real-world driving data and show improved performance over baseline approaches.

Ahmed Taha, Yi-Ting Chen, Teruhisa Misu et al.

We introduce an unsupervised formulation to estimate heteroscedastic uncertainty in retrieval systems. We propose an extension to triplet loss that models data uncertainty for each input. Besides improving performance, our formulation models local noise in the embedding space. It quantifies input uncertainty and thus enhances interpretability of the system. This helps identify noisy observations in query and search databases. Evaluation on both image and video retrieval applications highlight the utility of our approach. We highlight our efficiency in modeling local noise using two real-world datasets: Clothing1M and Honda Driving datasets. Qualitative results illustrate our ability in identifying confusing scenarios in various domains. Uncertainty learning also enables data cleaning by detecting noisy training labels.

Ahmed Taha, Yi-Ting Chen, Xitong Yang et al.

We cast visual retrieval as a regression problem by posing triplet loss as a regression loss. This enables epistemic uncertainty estimation using dropout as a Bayesian approximation framework in retrieval. Accordingly, Monte Carlo (MC) sampling is leveraged to boost retrieval performance. Our approach is evaluated on two applications: person re-identification and autonomous car driving. Comparable state-of-the-art results are achieved on multiple datasets for the former application. We leverage the Honda driving dataset (HDD) for autonomous car driving application. It provides multiple modalities and similarity notions for ego-motion action understanding. Hence, we present a multi-modal conditional retrieval network. It disentangles embeddings into separate representations to encode different similarities. This form of joint learning eliminates the need to train multiple independent networks without any performance degradation. Quantitative evaluation highlights our approach competence, achieving 6% improvement in a highly uncertain environment.

Vasili Ramanishka, Yi-Ting Chen, Teruhisa Misu et al.

Driving Scene understanding is a key ingredient for intelligent transportation systems. To achieve systems that can operate in a complex physical and social environment, they need to understand and learn how humans drive and interact with traffic scenes. We present the Honda Research Institute Driving Dataset (HDD), a challenging dataset to enable research on learning driver behavior in real-life environments. The dataset includes 104 hours of real human driving in the San Francisco Bay Area collected using an instrumented vehicle equipped with different sensors. We provide a detailed analysis of HDD with a comparison to other driving datasets. A novel annotation methodology is introduced to enable research on driver behavior understanding from untrimmed data sequences. As the first step, baseline algorithms for driver behavior detection are trained and tested to demonstrate the feasibility of the proposed task.