Kuan-Hsun Chen, Jing Li, F. Reghenzani, Jian-Jia Chen
{"title":"Introduction to the Special Issue on Fault-Resilient Cyber-Physical Systems – Part I","authors":"Kuan-Hsun Chen, Jing Li, F. Reghenzani, Jian-Jia Chen","doi":"10.1145/3677021","DOIUrl":"https://doi.org/10.1145/3677021","url":null,"abstract":"","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141670443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"ACM TCPS Foreword to Special Issue for ICCPS 2022","authors":"Sayan Mitra, N. Venkatasubramanian","doi":"10.1145/3661449","DOIUrl":"https://doi.org/10.1145/3661449","url":null,"abstract":"","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141004472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sepideh Safari, Shayan Shokri, S. Hessabi, Pejman Lotfi-Kamran
With the advent of multicore platforms in designing Mixed-Criticality Systems (MCSs), simultaneous management of reliability and energy while guaranteeing an acceptable service level for low-criticality tasks is a crucial challenge. To ensure the reliability of the MCSs against transient faults, fault-tolerant techniques are employed which will increase energy consumption. To mitigate the energy overhead, the Dynamic Voltage and Frequency Scaling (DVFS) technique will be exploited. However, this technique might lead to violating the timing constraints of high-criticality tasks. Therefore, this paper presents, for the first time, the low-energy checkpointing technique to guarantee the reliability of multiple preemptive periodic mixed-criticality tasks in a multicore platform. In contrast to the previous works in checkpointing technique which consider a specific number of faults that all the tasks in the system should tolerate, in this paper, the number of tolerable faults for each execution section of a task, and in each voltage and frequency level is determined through proposed formulas to meet the reliability target based on safety standards. Then, our proposed method determines the number of checkpoints and their non-uniform intervals for the normal and overrun sections of each task to reduce energy consumption, respectively. Moreover, the unified demand bound function (DBF) analysis is proposed for analyzing the schedulability of the task set, where each high-criticality task meets its timing and reliability constraints, and low-criticality tasks execute based on their derived guaranteed periods in each operational mode of the system. Experimental results show that our proposed scheme meets the timing and reliability constraints while at the same time, improving the QoS of low-criticality tasks, and managing energy consumption with an average of 29.49%, and 32.78%, respectively.
{"title":"LEC-MiCs: Low-Energy Checkpointing in Mixed-Criticality Multi-Core Systems","authors":"Sepideh Safari, Shayan Shokri, S. Hessabi, Pejman Lotfi-Kamran","doi":"10.1145/3653720","DOIUrl":"https://doi.org/10.1145/3653720","url":null,"abstract":"With the advent of multicore platforms in designing Mixed-Criticality Systems (MCSs), simultaneous management of reliability and energy while guaranteeing an acceptable service level for low-criticality tasks is a crucial challenge. To ensure the reliability of the MCSs against transient faults, fault-tolerant techniques are employed which will increase energy consumption. To mitigate the energy overhead, the Dynamic Voltage and Frequency Scaling (DVFS) technique will be exploited. However, this technique might lead to violating the timing constraints of high-criticality tasks. Therefore, this paper presents, for the first time, the low-energy checkpointing technique to guarantee the reliability of multiple preemptive periodic mixed-criticality tasks in a multicore platform. In contrast to the previous works in checkpointing technique which consider a specific number of faults that all the tasks in the system should tolerate, in this paper, the number of tolerable faults for each execution section of a task, and in each voltage and frequency level is determined through proposed formulas to meet the reliability target based on safety standards. Then, our proposed method determines the number of checkpoints and their non-uniform intervals for the normal and overrun sections of each task to reduce energy consumption, respectively. Moreover, the unified demand bound function (DBF) analysis is proposed for analyzing the schedulability of the task set, where each high-criticality task meets its timing and reliability constraints, and low-criticality tasks execute based on their derived guaranteed periods in each operational mode of the system. Experimental results show that our proposed scheme meets the timing and reliability constraints while at the same time, improving the QoS of low-criticality tasks, and managing energy consumption with an average of 29.49%, and 32.78%, respectively.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140379058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Social Internet of Vehicles (SIoV) is an emerging connected vehicular networking framework among specialized social vehicles to share and disseminate important information like traffic updates, weather conditions, parking slots, etc. This study aims to form an SIoV network among emergency vehicles for their frequent communication to improve the throughput, average waiting time, queue length, and speed during vehicular movement while crossing the intersection in the city. To address this, we propose a novel smart traffic light controller-assisted software-defined vehicular networking-enabled SIoV framework for emergency vehicles. Emergency vehicles form an SIoV network by utilizing Software-Defined Vehicular Networking (SDVN) architecture in Vehicle to Vehicle and Vehicle to Infrastructure communication. The SDVN module is used to offer two essential services: 1) SIoV-based road-lane prioritization, and 2) congestion prevention signal generation for the smart traffic light controller. An SDVN-MP algorithm is proposed to generate an effective traffic light control signal with an SDVN controller feedback signal. Furthermore, to improve the SIoV movement in the city, two levels of prioritization: 1) SIoV, and 2) the road lane with SIoV, are done. The first level of prioritization is to assign higher weightage to the social vehicular entities, and the second level is to prioritize the respective road lane based on SIoV quantity. The proposed framework is validated through a realistic simulation study on the Indian city OpenStreetMap utilizing the Simulation of Urban MObility simulator. The experimental findings demonstrate that the SDVN-MP model enhances (state-of-the-art) comparative performance by 22.5% to 55.2%, 1.2% to 82.7%, 1.6% to 38.4%, and 1.8% to 12.4% for average waiting time, average speed, average queue length, and average throughput metrics, respectively.
{"title":"SIoV Mobility Management using SDVN-enabled Traffic Light Cooperative Framework","authors":"Neetesh Kumar, Navjot Singh, Anuj Sachan, Rashmi Chaudhry","doi":"10.1145/3653721","DOIUrl":"https://doi.org/10.1145/3653721","url":null,"abstract":"Social Internet of Vehicles (SIoV) is an emerging connected vehicular networking framework among specialized social vehicles to share and disseminate important information like traffic updates, weather conditions, parking slots, etc. This study aims to form an SIoV network among emergency vehicles for their frequent communication to improve the throughput, average waiting time, queue length, and speed during vehicular movement while crossing the intersection in the city. To address this, we propose a novel smart traffic light controller-assisted software-defined vehicular networking-enabled SIoV framework for emergency vehicles. Emergency vehicles form an SIoV network by utilizing Software-Defined Vehicular Networking (SDVN) architecture in Vehicle to Vehicle and Vehicle to Infrastructure communication. The SDVN module is used to offer two essential services: 1) SIoV-based road-lane prioritization, and 2) congestion prevention signal generation for the smart traffic light controller. An SDVN-MP algorithm is proposed to generate an effective traffic light control signal with an SDVN controller feedback signal. Furthermore, to improve the SIoV movement in the city, two levels of prioritization: 1) SIoV, and 2) the road lane with SIoV, are done. The first level of prioritization is to assign higher weightage to the social vehicular entities, and the second level is to prioritize the respective road lane based on SIoV quantity. The proposed framework is validated through a realistic simulation study on the Indian city OpenStreetMap utilizing the Simulation of Urban MObility simulator. The experimental findings demonstrate that the SDVN-MP model enhances (state-of-the-art) comparative performance by 22.5% to 55.2%, 1.2% to 82.7%, 1.6% to 38.4%, and 1.8% to 12.4% for average waiting time, average speed, average queue length, and average throughput metrics, respectively.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140384208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Harini Sridhar, Gaojian Huang, Adam Thorpe, Meeko Oishi, Brandon J. Pitts
Partially autonomous driving systems often require the human driver to take control at any moment, yet by their design, often cause difficulty with attention management. In this preliminary study, we propose a data- and dynamics-driven approach to characterize driving performance in a partially autonomous vehicle during a manual braking event, under attentive or mind wandering states. A 10-participant experiment was completed in an advanced driving simulator. We employ a non-parametric learning technique, conditional distribution embeddings, to the driving simulator data, to evaluate likelihood of successfully completing the braking maneuver, under both attentive and mind wandering states. Our approach shows a statistically significant difference in braking profiles during mind wandering and non-mind wandering episodes for each participant. Our results reveal that heterogeneity in driving performance may have important implications for the design of autonomy that is responsive to attentional states. Data-driven tools, such as the one proposed here, may be useful in designing participant-specific alerts and warnings for control handovers and other safety-critical maneuvers, because of their potential to accommodate heterogeneous response.
{"title":"Characterizing the effect of mind wandering on partially autonomous braking dynamics","authors":"Harini Sridhar, Gaojian Huang, Adam Thorpe, Meeko Oishi, Brandon J. Pitts","doi":"10.1145/3653678","DOIUrl":"https://doi.org/10.1145/3653678","url":null,"abstract":"Partially autonomous driving systems often require the human driver to take control at any moment, yet by their design, often cause difficulty with attention management. In this preliminary study, we propose a data- and dynamics-driven approach to characterize driving performance in a partially autonomous vehicle during a manual braking event, under attentive or mind wandering states. A 10-participant experiment was completed in an advanced driving simulator. We employ a non-parametric learning technique, conditional distribution embeddings, to the driving simulator data, to evaluate likelihood of successfully completing the braking maneuver, under both attentive and mind wandering states. Our approach shows a statistically significant difference in braking profiles during mind wandering and non-mind wandering episodes for each participant. Our results reveal that heterogeneity in driving performance may have important implications for the design of autonomy that is responsive to attentional states. Data-driven tools, such as the one proposed here, may be useful in designing participant-specific alerts and warnings for control handovers and other safety-critical maneuvers, because of their potential to accommodate heterogeneous response.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140219844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Sulieman, Mengyu Liu, M. C. Gursoy, Fanxin Kong
Unmanned aerial vehicles (UAVs) have various applications in different settings, including e.g., surveillance, packet delivery, emergency response, data collection in the Internet of Things (IoT), and connectivity in cellular networks. However, this technology comes with many risks and challenges such as vulnerabilities to malicious cyber-physical attacks. This paper studies the problem of path planning for UAVs under GPS sensor permanent faults in a cyber-physical system (CPS) perspective. Based on studying and analyzing the CPS architecture of the UAV, the cyber “attacks and threats” are differentiated from attacks on sensors and communication components. An efficient way to address this problem is to introduce a novel approach for UAV’s path planning resilience to GPS permanent faults artificial potential field algorithm (RCA-APF). The proposed algorithm completes the three stages in a coordinated manner. In the first stage, the permanent faults on the GPS sensor of the UAV are detected, and the UAV starts to divert from its initial path planning. In the second stage, we estimated the location of the UAV under GPS permanent fault using Received Signal Strength (RSS) trilateration localization approach. In the final stage of the algorithm, we implemented the path planning of the UAV using an open-source UAV simulator. Experimental and simulation results demonstrate the performance of the algorithm and its effectiveness, resulting in efficient path planning for the UAV.
{"title":"Path Planning for UAVs Under GPS Permanent Faults","authors":"M. Sulieman, Mengyu Liu, M. C. Gursoy, Fanxin Kong","doi":"10.1145/3653074","DOIUrl":"https://doi.org/10.1145/3653074","url":null,"abstract":"Unmanned aerial vehicles (UAVs) have various applications in different settings, including e.g., surveillance, packet delivery, emergency response, data collection in the Internet of Things (IoT), and connectivity in cellular networks. However, this technology comes with many risks and challenges such as vulnerabilities to malicious cyber-physical attacks. This paper studies the problem of path planning for UAVs under GPS sensor permanent faults in a cyber-physical system (CPS) perspective. Based on studying and analyzing the CPS architecture of the UAV, the cyber “attacks and threats” are differentiated from attacks on sensors and communication components. An efficient way to address this problem is to introduce a novel approach for UAV’s path planning resilience to GPS permanent faults artificial potential field algorithm (RCA-APF). The proposed algorithm completes the three stages in a coordinated manner. In the first stage, the permanent faults on the GPS sensor of the UAV are detected, and the UAV starts to divert from its initial path planning. In the second stage, we estimated the location of the UAV under GPS permanent fault using Received Signal Strength (RSS) trilateration localization approach. In the final stage of the algorithm, we implemented the path planning of the UAV using an open-source UAV simulator. Experimental and simulation results demonstrate the performance of the algorithm and its effectiveness, resulting in efficient path planning for the UAV.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140227848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deep neural networks (DNNs) are effective tools for learning-enabled cyber-physical systems (CPSs) that handle high-dimensional image data. However, DNNs may make incorrect decisions when presented with inputs outside the distribution of their training data. These inputs can compromise the safety of CPSs. So, it becomes crucial to detect inputs as out-of-distribution (OOD) and interpret the reasons for their classification as OOD. In this study, we propose an interpretable learning method to detect OOD caused by meteorological features like darkness, lightness, and rain. To achieve this, we employ a variational autoencoder (VAE) to map high-dimensional image data to a lower-dimensional latent space. We then focus on a specific latent dimension and encourage it to classify different intensities of a particular meteorological feature in a monotonically increasing manner. This is accomplished by incorporating two additional terms into the VAE’s loss function: a classification loss and a positional loss. During training, we optimize the utilization of label information for classification. Remarkably, our results demonstrate that using only (25% ) of the training data labels is sufficient to train a single pre-selected latent dimension to classify different intensities of a specific meteorological feature. We evaluate the proposed method on two distinct datasets, CARLA and Duckietown, employing two different rain-generation methods. We show that our approach outperforms existing approaches by at least (15% ) in the F1 score and precision when trained and tested on CARLA dataset.
{"title":"Interpretable Latent Space for Meteorological Out-of-Distribution Detection via Weak Supervision","authors":"Suman Das, Michael Yuhas, Rachel Koh, A. Easwaran","doi":"10.1145/3651224","DOIUrl":"https://doi.org/10.1145/3651224","url":null,"abstract":"\u0000 Deep neural networks (DNNs) are effective tools for learning-enabled cyber-physical systems (CPSs) that handle high-dimensional image data. However, DNNs may make incorrect decisions when presented with inputs outside the distribution of their training data. These inputs can compromise the safety of CPSs. So, it becomes crucial to detect inputs as out-of-distribution (OOD) and interpret the reasons for their classification as OOD. In this study, we propose an interpretable learning method to detect OOD caused by meteorological features like darkness, lightness, and rain. To achieve this, we employ a variational autoencoder (VAE) to map high-dimensional image data to a lower-dimensional latent space. We then focus on a specific latent dimension and encourage it to classify different intensities of a particular meteorological feature in a monotonically increasing manner. This is accomplished by incorporating two additional terms into the VAE’s loss function: a classification loss and a positional loss. During training, we optimize the utilization of label information for classification. Remarkably, our results demonstrate that using only\u0000 \u0000 (25% )\u0000 \u0000 of the training data labels is sufficient to train a single pre-selected latent dimension to classify different intensities of a specific meteorological feature. We evaluate the proposed method on two distinct datasets, CARLA and Duckietown, employing two different rain-generation methods. We show that our approach outperforms existing approaches by at least\u0000 \u0000 (15% )\u0000 \u0000 in the\u0000 F1 score\u0000 and\u0000 precision\u0000 when trained and tested on CARLA dataset.\u0000","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140258321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Trier Mortlock, A. Malawade, Kohei Tsujio, M. A. Al Faruque
Autonomous vehicles are cyber-physical systems that combine embedded computing and deep learning with physical systems to perceive the world, predict future states, and safely control the vehicle through changing environments. The ability of an autonomous vehicle to accurately predict the motion of other road users across a wide range of diverse scenarios is critical for both motion planning and safety. However, existing motion prediction methods do not explicitly model contextual information about the environment, which can cause significant variations in performance across diverse driving scenarios. To address this limitation, we propose CASTNet : a dynamic, context-aware approach for motion prediction that (i) identifies the current driving context using a spatio-temporal model, (ii) adapts an ensemble of motion prediction models to fit the current context, and (iii) applies novel trajectory fusion methods to combine predictions output by the ensemble. This approach enables CASTNet to improve robustness by minimizing motion prediction error across diverse driving scenarios. CASTNet is highly modular and can be used with various existing image processing backbones and motion predictors. We demonstrate how CASTNet can improve both CNN-based and graph-learning-based motion prediction approaches and conduct ablation studies on the performance, latency, and model size for various ensemble architecture choices. In addition, we propose and evaluate several attention-based spatio-temporal models for context identification and ensemble selection. We also propose a modular trajectory fusion algorithm that effectively filters, clusters, and fuses the predicted trajectories output by the ensemble. On the nuScenes dataset, our approach demonstrates more robust and consistent performance across diverse, real-world driving contexts than state-of-the-art techniques.
{"title":"CASTNet: A Context-Aware, Spatio-Temporal Dynamic Motion Prediction Ensemble for Autonomous Driving","authors":"Trier Mortlock, A. Malawade, Kohei Tsujio, M. A. Al Faruque","doi":"10.1145/3648622","DOIUrl":"https://doi.org/10.1145/3648622","url":null,"abstract":"\u0000 Autonomous vehicles are cyber-physical systems that combine embedded computing and deep learning with physical systems to perceive the world, predict future states, and safely control the vehicle through changing environments. The ability of an autonomous vehicle to accurately predict the motion of other road users across a wide range of diverse scenarios is critical for both motion planning and safety. However, existing motion prediction methods do not explicitly model contextual information about the environment, which can cause significant variations in performance across diverse driving scenarios. To address this limitation, we propose\u0000 CASTNet\u0000 : a dynamic, context-aware approach for motion prediction that (i) identifies the current driving context using a spatio-temporal model, (ii) adapts an ensemble of motion prediction models to fit the current context, and (iii) applies novel trajectory fusion methods to combine predictions output by the ensemble. This approach enables CASTNet to improve robustness by minimizing motion prediction error across diverse driving scenarios. CASTNet is highly modular and can be used with various existing image processing backbones and motion predictors. We demonstrate how CASTNet can improve both CNN-based and graph-learning-based motion prediction approaches and conduct ablation studies on the performance, latency, and model size for various ensemble architecture choices. In addition, we propose and evaluate several attention-based spatio-temporal models for context identification and ensemble selection. We also propose a modular trajectory fusion algorithm that effectively filters, clusters, and fuses the predicted trajectories output by the ensemble. On the nuScenes dataset, our approach demonstrates more robust and consistent performance across diverse, real-world driving contexts than state-of-the-art techniques.\u0000","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140454641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Uncertainty in the predictions of learning-enabled components hinders their deployment in safety-critical cyber-physical systems (CPS). A shift from the training distribution of a learning-enabled component (LEC) is one source of uncertainty in the LEC’s predictions. Detection of this shift or out-of-distribution (OOD) detection on individual datapoints has therefore gained attention recently. But in many applications, inputs to CPS form a temporal sequence. Existing techniques for OOD detection in time-series data for CPS either do not exploit temporal relationships in the sequence or do not provide any guarantees on detection. We propose using deviation from the in-distribution temporal equivariance as the non-conformity measure in conformal anomaly detection framework for OOD detection in time-series data for CPS. Computing independent predictions from multiple conformal detectors based on the proposed measure and combining these predictions by Fisher’s method leads to the proposed detector CODiT with bounded false alarms. CODiT performs OOD detection on fixed-length windows of consecutive time-series datapoints by using Fisher value of the input window. We further propose performing OOD detection on real-time time-series traces of variable lengths with bounded false alarms. This can be done by using CODiT to compute Fisher values of the sliding windows in the input trace and combining these values by a merging function. Merging functions such as Harmonic Mean, Arithmetic Mean, Geometric Mean, and Bonferroni Method, etc. can be used to combine Fisher values of the sliding windows in the input trace, and the combined value can be used for OOD detection on the trace with bounded false alarm rate guarantees. We illustrate the efficacy of CODiT by achieving state-of-the-art results in two case studies for OOD detection on fixed-length windows. The first one is on an autonomous driving system with perception (or vision) LEC. The second case study is on a medical CPS for walking pattern or GAIT analysis where physiological (non-vision) data is collected with force-sensitive resistors attached to the subject’s body. For OOD detection on variable length traces, we consider the same case studies on the autonomous driving system and medical CPS for GAIT analysis. We report our results with four merging functions on the Fisher values computed by CODiT on the sliding windows of the input trace. We also compare the false alarm rate guarantees by these four merging functions in the autonomous driving system case study. Code, data, and trained models are available at https://github.com/kaustubhsridhar/time-series-OOD.
{"title":"Out-of-Distribution Detection in Dependent Data for Cyber-Physical Systems with Conformal Guarantees","authors":"Ramneet Kaur, Yahan Yang, O. Sokolsky, Insup Lee","doi":"10.1145/3648005","DOIUrl":"https://doi.org/10.1145/3648005","url":null,"abstract":"Uncertainty in the predictions of learning-enabled components hinders their deployment in safety-critical cyber-physical systems (CPS). A shift from the training distribution of a learning-enabled component (LEC) is one source of uncertainty in the LEC’s predictions. Detection of this shift or out-of-distribution (OOD) detection on individual datapoints has therefore gained attention recently. But in many applications, inputs to CPS form a temporal sequence. Existing techniques for OOD detection in time-series data for CPS either do not exploit temporal relationships in the sequence or do not provide any guarantees on detection. We propose using deviation from the in-distribution temporal equivariance as the non-conformity measure in conformal anomaly detection framework for OOD detection in time-series data for CPS. Computing independent predictions from multiple conformal detectors based on the proposed measure and combining these predictions by Fisher’s method leads to the proposed detector CODiT with bounded false alarms. CODiT performs OOD detection on fixed-length windows of consecutive time-series datapoints by using Fisher value of the input window. We further propose performing OOD detection on real-time time-series traces of variable lengths with bounded false alarms. This can be done by using CODiT to compute Fisher values of the sliding windows in the input trace and combining these values by a merging function. Merging functions such as Harmonic Mean, Arithmetic Mean, Geometric Mean, and Bonferroni Method, etc. can be used to combine Fisher values of the sliding windows in the input trace, and the combined value can be used for OOD detection on the trace with bounded false alarm rate guarantees.\u0000 We illustrate the efficacy of CODiT by achieving state-of-the-art results in two case studies for OOD detection on fixed-length windows. The first one is on an autonomous driving system with perception (or vision) LEC. The second case study is on a medical CPS for walking pattern or GAIT analysis where physiological (non-vision) data is collected with force-sensitive resistors attached to the subject’s body. For OOD detection on variable length traces, we consider the same case studies on the autonomous driving system and medical CPS for GAIT analysis. We report our results with four merging functions on the Fisher values computed by CODiT on the sliding windows of the input trace. We also compare the false alarm rate guarantees by these four merging functions in the autonomous driving system case study. Code, data, and trained models are available at https://github.com/kaustubhsridhar/time-series-OOD.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139781799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Uncertainty in the predictions of learning-enabled components hinders their deployment in safety-critical cyber-physical systems (CPS). A shift from the training distribution of a learning-enabled component (LEC) is one source of uncertainty in the LEC’s predictions. Detection of this shift or out-of-distribution (OOD) detection on individual datapoints has therefore gained attention recently. But in many applications, inputs to CPS form a temporal sequence. Existing techniques for OOD detection in time-series data for CPS either do not exploit temporal relationships in the sequence or do not provide any guarantees on detection. We propose using deviation from the in-distribution temporal equivariance as the non-conformity measure in conformal anomaly detection framework for OOD detection in time-series data for CPS. Computing independent predictions from multiple conformal detectors based on the proposed measure and combining these predictions by Fisher’s method leads to the proposed detector CODiT with bounded false alarms. CODiT performs OOD detection on fixed-length windows of consecutive time-series datapoints by using Fisher value of the input window. We further propose performing OOD detection on real-time time-series traces of variable lengths with bounded false alarms. This can be done by using CODiT to compute Fisher values of the sliding windows in the input trace and combining these values by a merging function. Merging functions such as Harmonic Mean, Arithmetic Mean, Geometric Mean, and Bonferroni Method, etc. can be used to combine Fisher values of the sliding windows in the input trace, and the combined value can be used for OOD detection on the trace with bounded false alarm rate guarantees. We illustrate the efficacy of CODiT by achieving state-of-the-art results in two case studies for OOD detection on fixed-length windows. The first one is on an autonomous driving system with perception (or vision) LEC. The second case study is on a medical CPS for walking pattern or GAIT analysis where physiological (non-vision) data is collected with force-sensitive resistors attached to the subject’s body. For OOD detection on variable length traces, we consider the same case studies on the autonomous driving system and medical CPS for GAIT analysis. We report our results with four merging functions on the Fisher values computed by CODiT on the sliding windows of the input trace. We also compare the false alarm rate guarantees by these four merging functions in the autonomous driving system case study. Code, data, and trained models are available at https://github.com/kaustubhsridhar/time-series-OOD.
{"title":"Out-of-Distribution Detection in Dependent Data for Cyber-Physical Systems with Conformal Guarantees","authors":"Ramneet Kaur, Yahan Yang, O. Sokolsky, Insup Lee","doi":"10.1145/3648005","DOIUrl":"https://doi.org/10.1145/3648005","url":null,"abstract":"Uncertainty in the predictions of learning-enabled components hinders their deployment in safety-critical cyber-physical systems (CPS). A shift from the training distribution of a learning-enabled component (LEC) is one source of uncertainty in the LEC’s predictions. Detection of this shift or out-of-distribution (OOD) detection on individual datapoints has therefore gained attention recently. But in many applications, inputs to CPS form a temporal sequence. Existing techniques for OOD detection in time-series data for CPS either do not exploit temporal relationships in the sequence or do not provide any guarantees on detection. We propose using deviation from the in-distribution temporal equivariance as the non-conformity measure in conformal anomaly detection framework for OOD detection in time-series data for CPS. Computing independent predictions from multiple conformal detectors based on the proposed measure and combining these predictions by Fisher’s method leads to the proposed detector CODiT with bounded false alarms. CODiT performs OOD detection on fixed-length windows of consecutive time-series datapoints by using Fisher value of the input window. We further propose performing OOD detection on real-time time-series traces of variable lengths with bounded false alarms. This can be done by using CODiT to compute Fisher values of the sliding windows in the input trace and combining these values by a merging function. Merging functions such as Harmonic Mean, Arithmetic Mean, Geometric Mean, and Bonferroni Method, etc. can be used to combine Fisher values of the sliding windows in the input trace, and the combined value can be used for OOD detection on the trace with bounded false alarm rate guarantees.\u0000 We illustrate the efficacy of CODiT by achieving state-of-the-art results in two case studies for OOD detection on fixed-length windows. The first one is on an autonomous driving system with perception (or vision) LEC. The second case study is on a medical CPS for walking pattern or GAIT analysis where physiological (non-vision) data is collected with force-sensitive resistors attached to the subject’s body. For OOD detection on variable length traces, we consider the same case studies on the autonomous driving system and medical CPS for GAIT analysis. We report our results with four merging functions on the Fisher values computed by CODiT on the sliding windows of the input trace. We also compare the false alarm rate guarantees by these four merging functions in the autonomous driving system case study. Code, data, and trained models are available at https://github.com/kaustubhsridhar/time-series-OOD.","PeriodicalId":505086,"journal":{"name":"ACM Transactions on Cyber-Physical Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139841698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：