Kernel-based traffic sign tracking to improve highway workzone recognition for reliable autonomous driving

Lee, Jong-Ho; Seo, Yongsok; Zhang, Wende; Wettergreen, David

doi:10.1109/itsc.2013.6728384

Cited by 10 publications

(8 citation statements)

References 21 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We developed a highway workzone recognition system to enable our self-driving car to respond to changes to highways' geometry and traffic rules [17], [28]. For example, we would like to see that our vehicle, while driving autonomously, is able to recognize a lowering of the speed limit due to road work and reduce its speed as posted.…”

Section: Methodsmentioning

confidence: 99%

“…Fig. 1 illustrates our workzone recognition system [17], [28]. Given an input image at t, our workzone recognition system begins with sign detection.…”

Section: Recognizing Temporary Highway Changesmentioning

confidence: 99%

“…To improve the performance of our color-based sign detection method, we develop an appearance-based sign-tracking method [16], [17]. Our sign tracker is motivated by the following two observations of a sign's appearance in a video.…”

Section: B Kernel-based Workzone Sign Trackingmentioning

confidence: 99%

See 2 more Smart Citations

Recognition of Highway Workzones for Reliable Autonomous Driving

Seo

Lee

Zhang

et al. 2014

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

In order to be deployed in real-world driving environments, self-driving cars must be able to recognize and respond to exceptional road conditions, such as highway workzones, because such unusual events can alter previously known traffic rules and road geometry. In this paper, we present a set of computer vision methods that recognize, through identification of workzone signs, the bounds of a highway workzone and temporary changes in highway driving environments. Through testing using video data about highway workzones recorded under various weather conditions, our approach was able to perfectly identify the boundaries of workzones and robustly detect a majority of driving condition changes. In addition to these tests, we evaluated, using a mock workzone setup, the usefulness of our workzone recognition systems' outputs for safe-guarding a self-driving car.Index Terms-Classification confidence propagation, highway workzone recognition, kernel-based sign tracking, learning color models for sign detection, sign classification.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Fig. 1 illustrates our workzone recognition system [17], [28]. Given an input image at t, our workzone recognition system begins with sign detection.…”

Section: Recognizing Temporary Highway Changesmentioning

confidence: 99%

See 1 more Smart Citation

Recognition of Highway Workzones for Reliable Autonomous Driving

Seo

Lee

Zhang

et al. 2014

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…We present a case study motivated by the software system of the self-driving car devel- oped at CMU [40]. Among various algorithmic tasks of the car, we chose a GPU accelerated version of the workzone recognition algorithm [31] (workzone) that periodically processes images collected from a camera. Two GPU-based matrix multiplication tasks (gpu matmul1 and gpu matmul2), and two CPU-bound tasks (cpu matmul1 and cpu matmul2) are also used to represent a subset of other tasks of the car.…”

Section: Practical Evaluationmentioning

confidence: 99%

A server-based approach for predictable GPU access with improved analysis

Kim

Patel

Wang

et al. 2018

Journal of Systems Architecture

View full text Add to dashboard Cite

We propose a server-based approach to manage a general-purpose graphics processing unit (GPU) in a predictable and efficient manner. Our proposed approach introduces a GPU server that is a dedicated task to handle GPU requests from other tasks on their behalf. The GPU server ensures bounded time to access the GPU, and allows other tasks to suspend during their GPU computation to save CPU cycles. By doing so, we address the two major limitations of the existing real-time synchronization-based GPU management approach: busy waiting within critical sections and long priority inversion. We have implemented a prototype of the server-based approach on a real embedded platform. This case study demonstrates the practicality and effectiveness of the server-based approach. Experimental results indicate that the server-based approach yields significant improvements in task schedulability over the existing synchronization-based approach in most practical settings. Although we focus on a GPU in this paper, the server-based approach can also be used for other types of computational accelerators. arXiv:1709.06613v2 [cs.DC] 11 May 2018 unit (GPU), which can greatly help in addressing the timing challenges of computation-intensive tasks by accelerating their execution.The use of GPUs in a time predictable manner brings up several challenges. First, many of today's commercial-off-the-shelf (COTS) GPUs do not support a preemption mechanism, and GPU access requests from application tasks are handled in a sequential, non-preemptive manner. This is primarily due to the high overhead expected on GPU context switching [39]. Although some recent GPU architectures, such as NVIDIA Pascal [2], claim to offer GPU preemption, there is no documentation regarding their explicit behavior, and existing drivers (and GPU programming APIs) do not offer any programmer control over GPU preemption at the time of writing this paper.Second, COTS GPU device drivers do not respect task priorities and the scheduling policy used in the system. Hence, in the worst case, the GPU access request of the highest-priority task may be delayed by the requests of all lower-priority tasks in the system, which could possibly cause unbounded priority inversion.The aforementioned issues have motivated the development of predictable GPU management techniques to ensure task timing constraints while achieving performance improvement [15,16,17,20,21,27,41]. Among them, the work in [15,16,17] introduces a synchronization-based approach that models GPUs as mutually-exclusive resources and uses real-time synchronization protocols to arbitrate GPU access. This approach has many benefits. First, it can schedule GPU requests from tasks in an analyzable manner, without making any change to GPU device drivers. Second, it allows the existing task schedulability analysis methods, originally developed for real-time synchronization protocols, to be easily applied to analyze tasks accessing GPUs. However, due to the underlying assumption on critical sections, this approach requires tasks t...

show abstract

“…The review of previous research and DOT practices show that light bar configurations on With the current Connected and Autonomous Vehicles, enhancing work zone detection is one way to reduce risk and improve safety Lee et al (2013). proposed a Kernel-based traffic sign tracking system to help improving work zone recognition by detecting and analyzing the shape…”

mentioning

confidence: 99%

Investigation of smart work zone technologies using mixed simulator and field studies

Si-yang¹

View full text Add to dashboard Cite

Safety is the top concern in transportation, especially in work zones, as work zones deviate from regular driving environment and driver behavior is very different. In order to protect workers and create a safer work zone environment, new technologies are proposed by agencies and deployed to work zones, however, some are without scientific study before deployment. Therefore, quantitative studies need to be conducted to show the effectiveness of technologies. Driving simulator is a safe and cost-effective way to test effectiveness of new designs and compare different configurations. Field study is another scientific way of testing, as it provides absolute validity, while simulator study provides relative validity. The synergy of field and simulator studies construct a precise experiment as field study calibrates simulator design and validates simulator results. Two main projects, Evaluation of Automated Flagger Assistance Devices (AFADs), and Evaluation of Green Lights on Truck-Mounted Attenuator (TMA), are discussed in this dissertation to illustrate the investigation of smart work zone technologies using mixed simulator and field studies, along with one simulator project investigating interaction between human driven car and autonomous truck platoon in work zones. Both field and simulator studies indicated that AFADs improved stationary work zone safety by enhancing visibility, isolating workers from immediate traffic, and conveying clear guidance message to traffic. The results of green light on TMAs implied an inverse relationship between visibility/awareness of work zone and arrow board recognition/easy on eyes, but did not show if any of the light configurations is superior. Results anticipated for autonomous truck platoon in work zones are drivers behave more uniformly after being educated about the meaning of signage displayed on the back of truck, and performance measured with signage would be more preferable than those without signage. Applications of statistics are extension of studies, including experimental design, survey design, and data analysis. Data obtained from AFAD and Green Light projects were utilized to illustrate the methodologies of data analysis and model building, which incorporated simulator data, biofeedback and survey response to interpret the relationship among driver perspective and mental status, and driving behavior. From the studies conducted, it could be concluded that mixed simulator and field study is a good fit for smart work zone technologies investigation. Simulators provide a safe environment, flexibility and cost-effectiveness, while field studies calibrate and validate simulator setup and its results. The collaboration of two forms of study generates legitimate and convincing results for investigations. Applying statistical methodologies into transportation simulator and field studies is a good way to make experiment and survey design more rational, and the statistical methods are applicable for further data analysis.

show abstract

Kernel-based traffic sign tracking to improve highway workzone recognition for reliable autonomous driving

Cited by 10 publications

References 21 publications

Recognition of Highway Workzones for Reliable Autonomous Driving

Recognition of Highway Workzones for Reliable Autonomous Driving

A server-based approach for predictable GPU access with improved analysis

Investigation of smart work zone technologies using mixed simulator and field studies

Contact Info

Product

Resources

About