Behavior Classification Research Articles

Cow Behavior Recognition Based on Wearable Nose Rings.

This study introduces a novel device designed to monitor dairy cow behavior, with a particular focus on feeding, rumination, and other behaviors. This study investigates the association between the cow behaviors and acceleration data collected using a three-axis, nose-mounted accelerometer, as well as the feasibility of improving the behavioral classification accuracy through machine learning. A total of 11 cows were used. We utilized three-axis acceleration sensors that were fixed to the cow's nose, and these devices provided detailed and unique data corresponding to their activity; in particular, a recorder was installed on each nasal device to obtain acceleration data, which were then used to calculate activity levels and changes. In addition, we visually observed the behavior of the cattle. The characteristic acceleration values during feeding, rumination, and other behavior were recorded; there were significant differences in the activity levels and changes between different behaviors. The results indicated that the nose ring device had the potential to accurately differentiate between eating and rumination behaviors, thus providing an effective method for the early detection of health problems and cattle management. The eating, rumination, and other behaviors of cows were classified with high accuracy using the machine learning technique, which can be used to calculate the activity levels and changes in cattle based on the data obtained from the nose-mounted, three-axis accelerometer.

Behavior classification and spatiotemporal analysis of grazing sheep using deep learning

Monitoring multiple behavioral patterns and grazing trajectories of individual sheep can provide valuable information for various aspects of livestock production. The emergence of low-cost miniature sensors, coupled with the continuous advancement of deep learning technologies, has ushered in a new generation of intelligent solutions for precision livestock farming. This study aims to explore the deployment methods of motion sensors, selection of data collection frequencies, and choice of deep learning algorithms to provide accurate classification of multiple behaviors in grazing sheep. Based on this, in conjunction with the acquired location information, the goal is to comprehend the spatiotemporal distribution of grazing sheep behaviors. Devices capable of collecting Inertial Measurement Unit (IMU) and location data were attached to the jaw, neck, and hind leg of sheep. Four datasets were created using IMU data with a frequency of 20 Hz and a 5 s time window from different positions (neck, neck & leg, jaw, jaw & leg). Two deep learning models, Convolutional Neural Network (CNN) - Long Short Term Memory (LSTM) and Temporal Convolutional Network (TCN)-Transformer, were employed to classify six grazing sheep behaviors: walking, standing, grazing, lying, standing-ruminating, and lying-ruminating. The results indicate that by fusing data from the neck- and leg-mounted devices and utilizing the CNN-LSTM model, the accuracy reached the highest at 99.3 %. Furthermore, a comparison was made regarding the behavior classification accuracy of this fused data at different IMU data frequencies (20, 10, 5, and 1 Hz). It was found that even when the data frequency was reduced to 1 Hz, the classification accuracy for the six sheep behaviors still exceeded 96 %. Additionally, the trained model with the highest accuracy was applied to monitoring grazing sheep behavior under two different management procedures. Further analysis of the time budgets of sheep behavior revealed differences in the durations of behaviors under different management procedures. For extensively grazing sheep, location information was also monitored and combined with behavior classification results to generate the spatiotemporal distribution of sheep behavior. The technologies presented are important for gaining further insights into the health status of grazing livestock and grassland, thereby enhancing grazing management practices by informing grazing choices and optimizing grassland utilization.

Optimized Meta-path extracted graph neural network for embedded computer performance evaluation model

The process of evaluating the effectiveness, speed, and general capabilities of a computer system that is embedded inside a larger device or system and created for a particular purpose is known as embedded computer performance evaluation. Specialized computing systems called embedded systems are built into a wide range of gadgets, including medical, industrial, , automotive and consumer electronics. In this manuscript, Optimized Meta-path extracted graph neural network for embedded computer performance evaluation model(MEGNN-EC-PEM) is proposed. Initially, the input data is obtained from real-time sensor measurements and system metrics for model training and testing. The input image is pre-processed using Orthogonal Master-slave Adaptive Notch Filter (OMANF) and it removes the noise from the collected data. Then, the pre- processed data are fed to embedded computer performance using Meta-path extracted graph neural network (MEGNN). In general, MEGNN does not express adapting optimization techniques to determine optimal parameters to assure precise embedded computer performance evaluation model. Hence, the Hunter–prey optimization algorithm (HPOA) is used to optimize Meta-path extracted graph neural network which accurately categorized embedded computer performance. Then, the proposed MEGNN-EC-PEM is implemented and the performance metrics like Modelling Accuracy, latency, Throughput, memory cost and Energy Consumption are analyzed. The performance of the MEGNN-EC-PEM approach attains 19.41%, 20.08% and 32.57% higher modelling accuracy, 22.41%, 23.08% and 24.57% lower latency and 23.01%, 23.08% and 24.07% higher throughput when analyzed through existing techniques like a state-based modelling approach for effective performance evaluation of embedded system architectures at transaction level (SMA-EPE-ESAT), Evaluating the performance of pre-trained convolutional neural network for audio classification on embedded schemes for anomaly detection in smart cities (PCNN-ACES-ADSC) and Animal behaviour classification via deep learning on embedded systems (ABC-VDL-ES) methods respectively.

Design a Deep Learning Model for Dairy Cow Abnormal Behaviour Prediction

When it comes to determining the health and well-being of cattle, it is critical to observe their behaviour. Another strong and cost-effective monitoring approach is a neural network-based monitoring system that analyses time series data from inertial sensors attached to cows. This is one of the most powerful and cost-effective monitoring systems currently available. Even though deep learning has made significant advances in pattern detection, large datasets are still required in many cases. When only a little amount of data is available, data augmentation is a very effective and low-cost pre-processing step for neural network-based systems. To study and analysis, a variety of ways for enhancing inertial sensor data based on cow behavioural data are discussed. Using convolutional neural networks, researchers are attempting to solve the difficult problem of identifying cow behaviour, which is made much more difficult by a scarcity of training data. Our proposed approaches areimproving the effectiveness of deep learning (CNN+LSTM) in the classification of cow behaviour while also lowering the total system costs associated with data collection and labelling.

A framework for ship abnormal behaviour detection and classification using AIS data

This paper proposes a method for detecting and classifying ship abnormal behaviour in ship trajectories. The method involves generating parameter profiles for the ship's trajectory and applying a Sliding Window algorithm to detect the ship's abnormal behaviour. Then, several features are adopted to effectively describe the characteristics of each ship's abnormal behaviour, such as the standard deviation of speed, detour factor, maximum drift angle, accumulative change of Course Over Ground and maximum lateral distance to the ship route. A density-based clustering algorithm is applied to group similar abnormal behaviour patterns according to the feature similarity, and the Random Forest Classification method is used to train a classification model based on the features extracted from the clusters. The proposed method is then tested on historical ship trajectory data provided by the Automatic Identification System. The results suggest that the method effectively identifies and classifies different abnormal behaviours in the ship trajectories.

Classification model of electricity consumption behavior based on sparse denoising autoencoder feature dimensionality reduction and spectral clustering

The development of electrical measurement technology has brought high latitude residential electricity consumption data to power companies, which contains the characteristics of users' electricity consumption behavior and provides data support for the behavior classification. In order to improve the efficiency of data feature extraction and the accuracy of electricity consumption behavior identification, a classification model based on sparse denoising autoencoder feature dimensionality reduction and spectral clustering is proposed in this paper. Firstly, the sparse denoising autoencoder (SDAE) and the manually defining electricity consumption characteristic indicators are deployed to extract features from the residential daily electricity consumption data, and then the spectral clustering is employed to classify the extracted electricity consumption characteristics. Secondly, the t-distributed stochastic neighbor embedding (t-SNE) is applied to visualize and analyze the classification results, and on this basis, the secondary classification is implemented to fix the issue of the confused electricity consumption behaviors. Finally, the typical consumption behavior curves are calculated by Gaussian distance weighting method, and the characteristics of power consumption behavior are analyzed and summarized. The proposed approach is evaluated and verified by using the electricity dataset in Fujian, China.

An Integrated Framework with ADD-LSTM and DeepLabCut for Dolphin Behavior Classification

Caring for dolphins is a delicate process that requires experienced caretakers to pay close attention to their behavioral characteristics. However, caretakers may sometimes lack experience or not be able to give their full attention, which can lead to misjudgment or oversight. To address these issues, a dolphin behavior analysis system has been designed to assist caretakers in making accurate assessments. This study utilized image preprocessing techniques to reduce sunlight reflection in the pool and enhance the outline of dolphins, making it easier to analyze their movements. The dolphins were divided into 11 key points using an open-source tool called DeepLabCut, which accurately helped mark various body parts for skeletal detection. The AquaAI Dolphin Decoder (ADD) was then used to analyze six dolphin behaviors. To improve behavior recognition accuracy, the long short-term memory (LSTM) neural network was introduced. The ADD and LSTM models were integrated to form the ADD-LSTM system. Several classification models, including unidirectional and bidirectional LSTM, GRU, and SVM, were compared. The results showed that the ADD module combined with a double-layer bidirectional LSTM method achieved high accuracy in dolphin behavior analysis. The accuracy rates for each behavior exceeded 90%.

Semantic learning and attention dynamics for behavioral classification in police narratives

The proactive identification of behavioral health incidents concerns from police reports is a critical yet underexplored area. The law enforcement officers provide follow-up services to improve community life by manually analyzing and identifying generated public narrative reports after the 911 incident calls. Therefore, automatically identifying these behavioral health calls from public narrative reports helps reduce the current manual labor-intensive identification process for law enforcement officers. In this work, we introduce a novel, multi-faceted approach that combines manual expert annotations, natural language processing (NLP), and cutting-edge machine learning strategies to classify and understand these incidents within police narratives efficiently. Our proposed method retrieves relevant narrative reports utilizing domain knowledge as behavioral health cues as terms/keywords. Our approach automatically extracts different categories of behavioral health cues by utilizing the limited domain knowledge enabled behavioral health terms using a cosine similarity-based thresholding approach. The behavioral health classification model employs an automatic attention-aware feature representation of terms/keywords, categories, and narrative reports to identify behavioral health cases with an accuracy of 85%. Extensive evaluation shows that our proposed model outperforms all the state-of-the-art models by approximately 4%.

Towards automatic detection and classification of swimming pectinids behaviour: first developments on great scallops (Pecten maximus)

ABSTRACT Over the past decades, valvometric techniques have been commonly used to record valve opening activities of bivalves. Various relationships with environmental variations have been elucidated through different types of metrics extracted from valvometric signals (e.g. valve opening, cyclicity, specific behaviours). Although automated data processing methods exist, many specific behaviours are still annotated manually. This study proposes an algorithm to detect and classify the behaviours performed by the great scallop (Pecten maximus) in two categories: jump-like (JL) behaviours and other behaviours (OBs). These two categories differ in the shape of their valvometric signal, JL being movements of high amplitudes associated with ‘displacement movements’ (rotation, swimming, jumping, flipping) and OB grouping all other movements of lower amplitudes (‘common movements’), such as partial closures, which are produced routinely. This algorithm has been developed and tested on 10 scallop valve opening time series recorded using fully autonomous valvometers based on the Hall effect principle. The algorithm detected 93.65% ± 5.5 of manually annotated behaviours produced by scallops, with a false detection rate of less than 6.3% ± 5.5. Classification performances vary according to the type of behaviour. JL behaviours and OBs were well classified at 83.72% ± 23.09 and 98.92% ± 1.80, respectively. Analysis of the algorithm's outputs, highlighting potential daily trends in the production of certain behaviours, shows their relevance for acquiring information on the biology of scallops. By providing an efficient and flexible detection and classification method, this study is a first step towards the automation of bivalve behaviour detection. This study also highlights the importance of simultaneously using Hall sensors and accelerometers to accurately classify the complex behaviours of mobile bivalves such as P. maximus.

Sex-specific nest attendance rhythm and foraging habitat use in a colony-breeding waterbird

Abstract In most colony-breeding species, biparental care during both egg incubation and chick-rearing is inevitable for successful reproduction, requiring parents to coordinate their nest attendance and foraging time. The extent to which the rhythm of nest attendance is adjusted to temporal and spatial variation in food availability is poorly understood. Here, we investigate whether the rhythm of nest attendance interacts with the spatial and temporal availability of foraging habitats in Eurasian spoonbills Platalea leucorodia breeding on Schiermonnikoog, a Dutch Wadden Sea barrier island. Spoonbills are tactile foragers that forage during both day and night in habitats of varying salinity. GPS-tracking combined with acceleration-based behavioral classification of 9 female and 13 male adult spoonbills between 2013 and 2019 revealed that, despite nearby foraging opportunities following a tidal rhythm, nest attendance followed a sex-specific diel rhythm. During incubation and chick-rearing, females attended the nest at night and foraged during the day, while males showed the reverse rhythm. Females made more and shorter foraging trips to, almost exclusively, nearby marine habitats, whereas the larger males often made long trips to forage in more distant freshwater habitats. Before and after breeding, females as well as males foraged primarily at night, suggesting that this is the preferred period of foraging for both sexes. Nevertheless, foraging habitat use remained sex-specific, being most likely explained by size-dependent foraging techniques. To conclude, the sex-specific rhythm of nest attendance is not shaped by the spatial and temporal availability of foraging habitats.

Automated Analysis Pipeline for Extracting Saccade, Pupil, and Blink Parameters Using Video-Based Eye Tracking.

The tremendous increase in the use of video-based eye tracking has made it possible to collect eye tracking data from thousands of participants. The traditional procedures for the manual detection and classification of saccades and for trial categorization (e.g., correct vs. incorrect) are not viable for the large datasets being collected. Additionally, video-based eye trackers allow for the analysis of pupil responses and blink behaviors. Here, we present a detailed description of our pipeline for collecting, storing, and cleaning data, as well as for organizing participant codes, which are fairly lab-specific but nonetheless, are important precursory steps in establishing standardized pipelines. More importantly, we also include descriptions of the automated detection and classification of saccades, blinks, "blincades" (blinks occurring during saccades), and boomerang saccades (two nearly simultaneous saccades in opposite directions where speed-based algorithms fail to split them), This is almost entirely task-agnostic and can be used on a wide variety of data. We additionally describe novel findings regarding post-saccadic oscillations and provide a method to achieve more accurate estimates for saccade end points. Lastly, we describe the automated behavior classification for the interleaved pro/anti-saccade task (IPAST), a task that probes voluntary and inhibitory control. This pipeline was evaluated using data collected from 592 human participants between 5 and 93 years of age, making it robust enough to handle large clinical patient datasets. In summary, this pipeline has been optimized to consistently handle large datasets obtained from diverse study cohorts (i.e., developmental, aging, clinical) and collected across multiple laboratory sites.

End-to-end deep learning approach to mouse behavior classification from cortex-wide calcium imaging.

Deep learning is a powerful tool for neural decoding, broadly applied to systems neuroscience and clinical studies. Interpretable and transparent models that can explain neural decoding for intended behaviors are crucial to identifying essential features of deep learning decoders in brain activity. In this study, we examine the performance of deep learning to classify mouse behavioral states from mesoscopic cortex-wide calcium imaging data. Our convolutional neural network (CNN)-based end-to-end decoder combined with recurrent neural network (RNN) classifies the behavioral states with high accuracy and robustness to individual differences on temporal scales of sub-seconds. Using the CNN-RNN decoder, we identify that the forelimb and hindlimb areas in the somatosensory cortex significantly contribute to behavioral classification. Our findings imply that the end-to-end approach has the potential to be an interpretable deep learning method with unbiased visualization of critical brain regions.

Correction to: Staked deep ensemble model for intruder behaviour detection and classification in cloud

Correction to: Staked deep ensemble model for intruder behaviour detection and classification in cloud

Three stage classification framework with ranking scheme for distracted driver detection using heuristic-assisted strategy

In recent days, the number of road accidents are greatly increased in world wide. The people around the world died due to the rapid road crush. The driver behavior must be detected to neglect the occurrence of road accidents. In this work, the three stage deep learning-based techniques are developed to detect the distracted driver action. The developed model is implemented in three phase to detect the distracted behavior of the driver during the driving process. In the first stage, the images are gathered from standard public resources regarding various drivers while driving. Next, the deep features retrieved using VGG16 and Inception and Xception networks. Then, the weighted fusedfusion is done by gathered features, where the weight parameter is optimized through the Adaptive Gradient-based Optimizer (AGBO) algorithm. Further, the driver behavior classification is done through a Deep Belief Network (DBN) with Radial Basis Functions Neural Networks (RBF), where the hyperparameters of DBN and RBF are optimized through the same AGBO algorithm. Finally, the first set of classified outcomes is attained from the images. In the second stage, the gathered images are fed to the feature extraction stage, where the texture features and motion features are gathered from the images. These collected features are fed to the same optimized DBN with RBF and the second set of classified outcomes with human behaviors has been attained.In the third stage, the gathered image fralgoriom the dataset is fed to the PCA for extracting the features. The gathered features are given to the same optimized DBN with the RBF classifier to get the classified outcome. Later, the classified outcomes from the three stages are fed to the ranking procedure for detecting the distracted driver action. When considering the evaluation indices, the suggested distracted driver prediction model attains 97.67% of accuracy, 97.68% of sensitivity, 97.67% of specificity, and 97.55% of precision value. Thus, the performance of the developed model in the distracted driver prediction is enhanced than the existing approaches.

Fabrication and Characterization of Electrically Conductive 3D Printable TPU/MWCNT Filaments for Strain Sensing in Large Deformation Conditions

AbstractThis study investigates the development of thermoplastic polyurethane (TPU) filaments incorporating multi‐walled carbon nanotubes (MWCNT) to enhance strain‐sensing capabilities. Various MWCNT reinforcement ratios are used to produce customized feedstock for fused filament fabrication (FFF) 3D printing. Mechanical properties and the piezoresistive response of samples printed with these multifunctional filaments are concurrently evaluated. Surface morphology and microstructural observations reveal that higher MWCNT weight percentages increase filament surface roughness and rigidity. The microstructural modifications directly influence the tensile strength and strain energy of the printed samples. The study identifies an apparent percolation threshold within the 10–12 wt.% MWCNT range, indicating the formation of a conductive network. At this threshold, higher gauge factors are achieved at lower strains. A newly introduced Electro‐Mechanical Sensitivity Ratio (ESR) parameter enables the classification of composite behaviors into two distinct zones, offering the ability to tailor self‐sensing structures with on‐demand properties. Finally, flexible structures with proven application in soft robotics and shape morphing are fabricated and tested at different loading conditions to demonstrate the potential applicability of the custom filaments produced. The results highlight a pronounced piezoresistive response and superior load‐bearing performance in the examined structures.

Secure Enhancement for MQTT Protocol Using Distributed Machine Learning Framework.

The Message Queuing Telemetry Transport (MQTT) protocol stands out as one of the foremost and widely recognized messaging protocols in the field. It is often used to transfer and manage data between devices and is extensively employed for applications ranging from smart homes and industrial automation to healthcare and transportation systems. However, it lacks built-in security features, thereby making it vulnerable to many types of attacks such as man-in-the-middle (MitM), buffer overflow, pre-shared key, brute force authentication, malformed data, distributed denial-of-service (DDoS) attacks, and MQTT publish flood attacks. Traditional methods for detecting MQTT attacks, such as deep neural networks (DNNs), k-nearest neighbor (KNN), linear discriminant analysis (LDA), and fuzzy logic, may exist. The increasing prevalence of device connectivity, sensor usage, and environmental scalability become the most challenging aspects that novel detection approaches need to address. This paper presents a new solution that leverages an H2O-based distributed machine learning (ML) framework to improve the security of the MQTT protocol in networks, particularly in IoT environments. The proposed approach leverages the strengths of the H2O algorithm and architecture to enable real-time monitoring and distributed detection and classification of anomalous behavior (deviations from expected activity patterns). By harnessing H2O's algorithms, the identification and timely mitigation of potential security threats are achieved. Various H2O algorithms, including random forests, generalized linear models (GLMs), gradient boosting machine (GBM), XGBoost, and the deep learning (DL) algorithm, have been assessed to determine the most reliable algorithm in terms of detection performance. This study encompasses the development of the proposed algorithm, including implementation details and evaluation results. To assess the proposed model, various evaluation metrics such as mean squared error (MSE), root-mean-square error (RMSE), mean per class error (MCE), and log loss are employed. The results obtained indicate that the H2OXGBoost algorithm outperforms other H2O models in terms of accuracy. This research contributes to the advancement of secure IoT networks and offers a practical approach to enhancing the security of MQTT communication channels through distributed detection and classification techniques.

Large nesting expression in deer mice remains stable under conditions of visual deprivation despite heightened limbic involvement: Perspectives on compulsive-like behavior.

Visual stimuli and limbic activation varyingly influence obsessive-compulsive symptom expression and so impact treatment outcomes. Some symptom phenotypes, for example, covert repugnant thoughts, are likely less sensitive to sensory stimuli compared to symptoms with an extrinsic focus, that is, symptoms related to contamination, safety, and "just-right-perceptions." Toward an improved understanding of the neurocognitive underpinnings of obsessive-compulsive psychobiology, work in naturalistic animal model systems is useful. Here, we explored the impact of visual feedback and limbic processes on 24 normal (NNB) and large (LNB) nesting deer mice, respectively (as far as possible, equally distributed between sexes). Briefly, after behavioral classification into either the NNB or LNB cohorts, mice of each cohort were separated into two groups each and assessed for nesting expression under either standard light conditions or conditions of complete visual deprivation (VD). Nesting outcomes were assessed in terms of size and neatness. After nesting assessment completion, mice were euthanized, and samples of frontal-cortical and hippocampal tissues were collected to determine serotonin and noradrenaline concentrations. Our results show that LNB, as opposed to NNB, represents an inflexible and excessive behavioral phenotype that is not dependent on visually guided action-outcome processing, and that it associates with increased frontal-cortical and hippocampal noradrenaline concentrations, irrespective of lighting condition. Collectively, the current results are informing of the neurocognitive underpinnings of nesting behavior. It also provides a valuable foundation for continued investigations into the noradrenergic mechanisms that may influence the development and promulgation of excessive, rigid, and inflexible behaviors.

Deep learning-based classification of cattle behavior using accelerometer sensors

<span lang="EN-US">The increasing demand for food has led to the adoption of precision livestock, which relies on information and communication technology to promote the best practices in meat production. By automating various aspects of the industry, precision livestock allows for increased productivity, more effective management strategies, and decision-making. The paper proposes a methodology that uses deep learning techniques to automatically classify cattle behavior using accelerometer sensors embedded in collars. The work aims to enhance the efficiency and productivity of the industry by improving the classification of cattle behaviors, which is essential for farmers and barn managers to make informed decisions. We tested three different classification techniques to classify rumination, movement, resting, feeding, salting and other cattle behaviors and we achieved promising results that can contribute to a better understanding and management of cattle behavior in the livestock industry.</span>

Assessing Spontaneous Categorical Processing of Visual Shapes via Frequency-Tagging EEG.

Categorization is an essential cognitive and perceptual process, which happens spontaneously. However, earlier research often neglected the spontaneous nature of this process by mainly adopting explicit tasks in behavioral or neuroimaging paradigms. Here, we use frequency-tagging (FT) during electroencephalography (EEG) in 22 healthy human participants (both male and female) as a direct approach to pinpoint spontaneous visual categorical processing. Starting from schematic natural visual stimuli, we created morph sequences comprising 11 equal steps. Mirroring a behavioral categorical perception discrimination paradigm, we administered a FT-EEG oddball paradigm, assessing neural sensitivity for equally sized differences within and between stimulus categories. Likewise, mirroring a behavioral category classification paradigm, we administered a sweep FT-EEG oddball paradigm, sweeping from one end of the morph sequence to the other, thereby allowing us to objectively pinpoint the neural category boundary. We found that FT-EEG can implicitly measure categorical processing and discrimination. More specifically, we could derive an objective neural index of the required level to differentiate between the two categories, and this neural index showed the typical marker of categorical perception (i.e., stronger discrimination across as compared with within categories). The neural findings of the implicit paradigms were also validated using an explicit behavioral task. These results provide evidence that FT-EEG can be used as an objective tool to measure discrimination and categorization and that the human brain inherently and spontaneously (without any conscious or decisional processes) uses higher-level meaningful categorization information to interpret ambiguous (morph) shapes.

Stable training via elastic adaptive deep reinforcement learning for autonomous navigation of intelligent vehicles

The uncertain stability of deep reinforcement learning training on complex tasks impedes its development and deployment, especially in intelligent vehicles, such as intelligent surface vessels and self-driving cars. Complex and varied environmental states puzzle training of decision-making networks. Here we propose an elastic adaptive deep reinforcement learning algorithm to address these challenges and achieve autonomous navigation in intelligent vehicles. Our method trains the decision-making network over the function and optimization learning stages, in which the state space and action space of autonomous navigation tasks are pruned by choosing classic states and actions to reduce data similarity, facilitating more stable training. We introduce a task-adaptive observed behaviour classification technique in the function learning stage to divide state and action spaces into subspaces and identify classic states and actions. In which the classic states and actions are accumulated as the training dataset that enhances its training efficiency. In the subsequent optimization learning stage, the decision-making network is refined through meticulous exploration and accumulation of datasets. The proposed elastic adaptive deep reinforcement learning enables the decision-making network to effectively learn from complex state and action spaces, leading to more efficient training compared to traditional deep reinforcement learning approaches. Simulation results demonstrate the remarkable effectiveness of our method in training decision-making networks for intelligent vehicles. The findings validate that our method provides reliable and efficient training for decision-making networks in intelligent vehicles. Moreover, our method exhibits stability in training other tasks characterized by continuous state and action spaces.