Dynamic Risk Assessment for Vehicles of Higher Automation Levels by Deep Learning

Feth, Patrik; Akram, Mohammed Naveed; Schuster, René; Wasenmüller, Oliver

doi:10.1007/978-3-319-99229-7_48

Cited by 10 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the dynamic environment of mill yard operations, machines and vehicles work in quite close proximity, and in some cases, workers and site visitors may be moving around the operation area, which increases the need to constantly consider 360 degrees around the work area and complicates safety issues, especially when planning changes to work processes [15]. Furthermore, the yards have different road types (structured, unstructured, paved, unpaved) and are usually outdoors, and consequently affected by changing weather conditions (rain, snow, fog, drought, high heat).…”

Section: Research Motivation and Backgroundmentioning

confidence: 99%

Toward Autonomous Vehicles and Machinery in Mill Yards of the Forest Industry: Technologies and Proposals for Autonomous Vehicle Operations

et al. 2022

View full text Add to dashboard Cite

The use of autonomous systems on the wood processing sites of forest industries can significantly increase safety, productivity, and efficiency by reducing the number of monotonous and dangerous tasks conducted by human labor utilizing heavy machinery. However, autonomous machines are challenging and difficult to implement in the mill yards environment because of the dynamic and complex working environment, and partly unstructured processes. The inherent complexity of wood handling and storage tasks is requiring significant human expertise. Fast advancements in sensor technologies, machine learning techniques, and increases in the computational power enable progress in automated operation framework and algorithms development. This opens the door to introduce novel autonomous systems into this environment. With the aim of gaining a better understanding of current issues and facilitating optimal strategies for deploying high-level autonomous systems for the mill yard environments, this study: (1) Utilized a systematic literature review to map current autonomous technologies and algorithms suitable for adoption by the forest industry in automation of vehicles working in mill yards; (2) Summarizes and discusses the potential feasibility of the considered sensors and systems, adaption strategy, and high-level autonomous machinery implementation challenges in the mill yard environment; and (3) Proposes a system framework that integrates multiple technologies to enable autonomous navigation and material handling in mill yards. The study is the first of its kind as a comprehensive study on autonomous vehicles and machinery in mill yard environments. Our novel framework aids in the identification of follow-up research areas and thus promotes the adoption and use of complex autonomous systems in industrial environments.

show abstract

Section: Research Motivation and Backgroundmentioning

confidence: 99%

Toward Autonomous Vehicles and Machinery in Mill Yards of the Forest Industry: Technologies and Proposals for Autonomous Vehicle Operations

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Existing methods of assessment include AV focusing on risk-related figures and attributions. These methods consist of process-driven [4,[17][18][19], probability [26][27][28][29], model-based techniques [30][31][32], artificial intelligence methodologies [33][34][35] and lastly cooperative mode based approaches [36][37][38][39][40][41][42][43][44].…”

Section: Related Workmentioning

confidence: 99%

Infrastructure-Based Pedestrian Risk Tagging Methodology to Support AV Risk Assessment

et al. 2022

View full text Add to dashboard Cite

show abstract

“…They pioneered the notion of object importance in object detection. They asked human drivers to annotate the importance of objects in driving scenes and used the annotations to showcase the advantages of importance Neural networks to calculate the threat or risk of different scenes have also been explored [7], [8]. The main feature is that these DNN can estimate the risk of a situation based on visual input.…”

Section: Related Workmentioning

confidence: 99%

Risk Ranked Recall: Collision Safety Metric for Object Detection Systems in Autonomous Vehicles

Bansal

Singh

Verucchi

et al. 2021

2021 10th Mediterranean Conference on Embedded Computing (MECO)

View full text Add to dashboard Cite

Commonly used metrics for evaluation of object detection systems (precision, recall, mAP) do not give complete information about their suitability of use in safety critical tasks, like obstacle detection for collision avoidance in Autonomous Vehicles (AV). This work introduces the Risk Ranked Recall (R 3 ) metrics for object detection systems. The R 3 metrics categorize objects within three ranks. Ranks are assigned based on an objective cyber-physical model for the risk of collision. Recall is measured for each rank.

show abstract

Dynamic Risk Assessment for Vehicles of Higher Automation Levels by Deep Learning

Cited by 10 publications

References 17 publications

Toward Autonomous Vehicles and Machinery in Mill Yards of the Forest Industry: Technologies and Proposals for Autonomous Vehicle Operations

Toward Autonomous Vehicles and Machinery in Mill Yards of the Forest Industry: Technologies and Proposals for Autonomous Vehicle Operations

Infrastructure-Based Pedestrian Risk Tagging Methodology to Support AV Risk Assessment

Risk Ranked Recall: Collision Safety Metric for Object Detection Systems in Autonomous Vehicles

Contact Info

Product

Resources

About