Risk Behavior-Based Trajectory Prediction for Construction Site Safety Monitoring

Rashid, Khandakar M.; Behzadan, Amir H.

doi:10.1061/(asce)co.1943-7862.0001420

Cited by 35 publications

(17 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Warning generation based on proximity to equipment hazards [11] and to static hazards [13] GPS localization is appropriate for tracking resources outdoors.…”

Section: Applicability To Current Research Problemmentioning

confidence: 99%

“…To address these human-related shortcomings, there has been much interest in leveraging various technologies such as virtual reality and sensors to automate and augment human hazard recognition and avoidance. For example, recent research has focused on developing various hazard proximity warning systems using localization systems [9][10][11][12] that alert workers when they get dangerously close to safety hazards that can cause injury. These efforts have demonstrated that workers are able to adopt appropriate hazard avoidance measures after being notified of their proximity to safety hazards such as mobile equipment [13,14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Incorporating Worker Awareness in the Generation of Hazard Proximity Warnings

Chan

Louis

Albert

2020

Sensors

View full text Add to dashboard Cite

Proximity warning systems for construction sites do not consider whether workers are already aware of the hazard prior to issuing warnings. This can generate redundant and distracting alarms that interfere with worker ability to adopt timely and appropriate avoidance measures; and cause alarm fatigue, which instigates workers to habitually disable the system or ignore the alarms; thereby increasing the risk of injury. Thus, this paper integrates the field-of-view of workers as a proxy for hazard awareness to develop an improved hazard proximity warning system for construction sites. The research first developed a rule-based model for the warning generation, which was followed by a virtual experiment to evaluate the integration of worker field-of-view in alarm generation. Based on these findings, an improved hazard proximity warning system incorporating worker field-of-view was developed for field applications that utilizes wearable inertial measurement units and localization sensors. The system’s effectiveness is illustrated through several case studies. This research provides a fresh perspective to the growing adoption of wearable sensors by incorporating the awareness of workers into the generation of hazard alarms. The proposed system is anticipated to reduce unnecessary and distracting alarms which can potentially lead to superior safety performance in construction.

show abstract

“…Warning generation based on proximity to equipment hazards [11] and to static hazards [13] GPS localization is appropriate for tracking resources outdoors.…”

Section: Applicability To Current Research Problemmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Incorporating Worker Awareness in the Generation of Hazard Proximity Warnings

Chan

Louis

Albert

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Numerous data augmentation techniques have been tried and tested in computer vision [24][25][26][27][28] and speech recognition domain [29][30][31]. Charalambous and Bharath (2016) [24] introduced a simulation-based methodology which can be used for generating synthetic video data and sequence for machine/deep learning gait recognition algorithms.…”

Section: Data Augmentation For Classificationmentioning

confidence: 99%

“…in order to generate synthetic data for training and testing machine/deep learning algorithms [25,26,30]. Moreover, in the speech recognition domain, studies have applied techniques such as vocal tract length normalization [27][28], speech rate, and frequency-axis random distortion [27], label-preserving audio transformation [29] to improve the performance of learning algorithms.…”

Section: Data Augmentation For Classificationmentioning

confidence: 99%

Time-Warping: A Time Series Data Augmentation of IMU Data for Construction Equipment Activity Identification

Rashid¹,

Louis²

2019

Proceedings of the International Symposium on Automation and Robotics in Construction (IAARC)

Self Cite

View full text Add to dashboard Cite

Automated, real-time, and reliable equipment activity identification on construction sites can help to minimize idle times, improve operational efficiencies, and reduce emissions. Many previous efforts in activity identification have explored different machine learning algorithms that use time-series sensor data collected from inertial measurement units mounted on the equipment. However, machine learning algorithms requires large volume of training data collection from the field, as inadequate and smaller amounts of data results in model overfitting. This study proposes an automatic and real-time activity recognition framework by using data from multiple IMUs attached to equipment's moving and articulated parts. In doing so, first a time-series data augmentation technique called window-warping (WW) is introduced to generate synthetic training data from a smaller volume of field-collected data. Two supervised machine learning algorithms, artificial neural network (ANN), and K-nearest neighbour (KNN) were trained and evaluated using the augmented training data to identify equipment activity. The developed data augmentation methodology is validated using a case study of an earthmoving excavator. The results show the potential for using time-series data augmentation in training machine learning algorithms for construction equipment activity recognition using minimal data collected from the field.

show abstract

“…Automated activity identification of construction equipment enables real-time applications including, but not limited to, real-time productivity monitoring (Ahn et al, 2015;Louis and Dunston, 2016b;Kim H. et al, 2018), real-time safety analysis (Carbonari et al, 2011;Cheng and Teizer, 2013;Rashid et al, 2017;Rashid and Behzadan, 2018), real-time routing of resources (Louis and Dunston, 2016a), and realtime AR/VR visualization (Ku et al, 2011;Dong and Kamat, 2013;Park et al, 2014;You et al, 2018). Offline applications enabled by automated activity identification include preparing dynamic simulation input (Akhavian and Behzadan, 2013;Vahdatikhaki and Hammad, 2014), automated cycle time analysis (Vahdatikhaki and Hammad, 2014;Mathur et al, 2015;Kim H. et al, 2018), operation analysis (Vahdatikhaki and Hammad, 2014;, and fuel use analysis (Lewis et al, 2011) etc.…”

Section: Introductionmentioning

confidence: 99%

Automated Activity Identification for Construction Equipment Using Motion Data From Articulated Members

Rashid

Louis

2020

Front. Built Environ.

Self Cite

View full text Add to dashboard Cite

Real-time monitoring and assessment of construction resources have always been a challenge due to the unique, dynamic, and complex nature of each construction site and operation. The ability to automatically classify activities performed by various equipment in real time can aid in making timely tactical operational decisions that can lead to increased fleet productivity, reduced time and cost of operations, and minimized idle times. Such endeavors have traditionally been performed manually through human observation, making it extremely labor and time-intensive. Meanwhile, the development of low-cost micro-electro-mechanical systems (MEMS) with rapidly evolving computing, networking, and storage capabilities, along with the advances in computational techniques such as machine learning present new opportunities in the real-time activity identification domain. Even though previous studies have shown promising results for equipment activity identification at limited levels of detail, they have a fundamental limitation in their reliance on the equipment vibration. Equipment vibration is highly dependent on factors that are extrinsic to the performance of an activity itself such as ground conditions, age and condition of equipment, and operator skill. This aspect of current methods necessitates the collection of training data from the specific equipment of interest and requires manual labeling of training data, thereby limiting its application across different types of equipment and operating conditions. This paper investigates the use of activity-specific equipment motions instead of vibration for activity identification. This approach is the first step toward the larger goal of generating training data automatically from virtual kinematic models of equipment in the future. This paper also provides an array of sensitivity analyses in order to determine the most appropriate parameters for implementing machine learning algorithms for equipment activity identification. A case study was performed using an excavator working on an earthmoving site that demonstrated a significant improvement in equipment activity identification results by utilizing inertial measurement unit (IMU) data of different articulated elements over previous efforts. The results of this paper indicate that more accurate results for activity identification can be obtained by using articulated equipment motion over vibration, which paves the way for the automatic generation of labeled training data in the future.

show abstract

Risk Behavior-Based Trajectory Prediction for Construction Site Safety Monitoring

Cited by 35 publications

References 33 publications

Incorporating Worker Awareness in the Generation of Hazard Proximity Warnings

Incorporating Worker Awareness in the Generation of Hazard Proximity Warnings

Time-Warping: A Time Series Data Augmentation of IMU Data for Construction Equipment Activity Identification

Automated Activity Identification for Construction Equipment Using Motion Data From Articulated Members

Contact Info

Product

Resources

About