Nowadays, autonomous vehicle is an active research area, especially after the emergence of machine vision tasks with deep learning. In such a visual navigation system for autonomous vehicle, the controller captures images and predicts information so that the autonomous vehicle can safely navigate. In this paper, we first introduced small and medium-sized obstacles that were intentionally or unintentionally left on the road, which can pose hazards for both autonomous and human driving situations. Then, we discuss Markov random field (MRF) model by fusing three potentials (gradient potential, curvature prior potential, and depth variance potential) to segment the obstacles and non-obstacles into the hazardous environment. Since the segment of obstacles is done by MRF model, we can predict the information to safely navigate the autonomous vehicle form hazardous environment on the roadway by DNN model. We found that our proposed method can segment the obstacles accuracy from the blended background road and improve the navigation skills of the autonomous vehicle.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7506726 | PMC |
| http://dx.doi.org/10.3390/s20174719 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Through-the-sensor sub-bottom imaging using the self-noise of an autonomous underwater vehiclea).
JASA Express Lett
January 2025
Scripps Institution of Oceanography, University of California San Diego, La Jolla, California 92093,
This work demonstrates the feasibility of performing through-the-sensor (TTS) sub-bottom imaging using low-frequency ([100 Hz-1kHz]) self-noise generated by the propulsion of an autonomous underwater vehicle (AUV) acting as a source of opportunity. The self-noise was recorded by a short towed horizontal line array (11.4 m aperture) by the same AUV while it operated ∼35 m above the seabed along a range-dependent section at the New England shelf break.
View Article and Find Full Text PDFBroadband wireless battery-free acoustic identification tags for high data-rate underwater backscatter communication.
J Acoust Soc Am
January 2025
George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA.
Developing persistent and smart underwater markers is critical for improving navigation accuracy and communication capabilities of autonomous underwater vehicles (AUVs). A wireless acoustic identification tag, which uses a piezoelectric transducer tuned in the broadband ultrasonic range (200-500 kHz), was experimentally demonstrated to achieve highly efficient power transfer (source-to-tag electrical power efficiency of >2% at 6 m) and concurrent high data rate and backscatter level communication (>83.3 kbit s-1, >170 dB sound pressure level at 6 m) with potential operating range ≈ 10 m based on analytical extrapolations.
View Article and Find Full Text PDFEnhancing semantic segmentation for autonomous vehicle scene understanding in indian context using modified CANet model.
MethodsX
June 2025
Department of Computer Engineering, Pimpri Chinchwad College of Engineering, Nigdi, Pune 411044, India.
Recent advancements in artificial intelligence (AI) have increased interest in intelligent transportation systems, particularly autonomous vehicles. Safe navigation in traffic-heavy environments requires accurate road scene segmentation, yet traditional computer vision methods struggle with complex scenarios. This study emphasizes the role of deep learning in improving semantic segmentation using datasets like the Indian Driving Dataset (IDD), which presents unique challenges in chaotic road conditions.
View Article and Find Full Text PDFAggregate global features into separable hierarchical lane detection transformer.
Sci Rep
January 2025
School of Information Engineering, Henan University of Science and Technology, Luoyang, 471000, China.
Lane detection is one of the key functions to ensure the safe driving of autonomous vehicles, and it is a challenging task. In real driving scenarios, external factors inevitably interfere with the lane detection system, such as missing lane markings, harsh weather conditions, and vehicle occlusion. To enhance the accuracy and detection speed of lane detection in complex road environments, this paper proposes an end-to-end lane detection model with a pure Transformer architecture, which exhibits excellent detection performance in complex road scenes.
View Article and Find Full Text PDFVehicle speed measurement method using monocular cameras.
Sci Rep
January 2025
Computer and Information Engineering College, Inner Mongolia Agricultural University, Hohhot, 010000, China.
This paper proposes a method for fast and accurate vehicle speed measurement based on a monocular camera. Firstly, by establishing a new camera imaging model, the calibration method for variable focal lengths is optimized, simplifying the transformation process between the four coordinate systems in traditional camera imaging models, and the method does not need to restore the pixel coordinates to dedistortion. Secondly, based on the camera imaging model, a two-dimensional positioning algorithm is proposed.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!