scholarly journals Identification of Road-Surface Type Using Deep Neural Networks for Friction Coefficient Estimation

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 612 ◽  
Author(s):  
Eldar Šabanovič ◽  
Vidas Žuraulis ◽  
Olegas Prentkovskis ◽  
Viktor Skrickij
Keyword(s):  
Neural Networks ◽  
Friction Coefficient ◽  
Deep Neural Networks ◽  
Image Sensor ◽  
Video Data ◽  
Road Surface ◽  
Driver Assistance Systems ◽  
Surface Type ◽  
Braking Performance ◽  
The Road

Nowadays, vehicles have advanced driver-assistance systems which help to improve vehicle safety and save the lives of drivers, passengers and pedestrians. Identification of the road-surface type and condition in real time using a video image sensor, can increase the effectiveness of such systems significantly, especially when adapting it for braking and stability-related solutions. This paper contributes to the development of the new efficient engineering solution aimed at improving vehicle dynamics control via the anti-lock braking system (ABS) by estimating friction coefficient using video data. The experimental research on three different road surface types in dry and wet conditions has been carried out and braking performance was established with a car mathematical model (MM). Testing of a deep neural networks (DNN)-based road-surface and conditions classification algorithm revealed that this is the most promising approach for this task. The research has shown that the proposed solution increases the performance of ABS with a rule-based control strategy.

Research on Vehicle ABS Detection Technology Based on Flywheel Inertia Simulation

2013 ◽  
Vol 765-767 ◽  
pp. 2117-2122 ◽  
Author(s):  
Kai Zhou ◽  
Ri Sha Na ◽  
Xu Dong Wang
Keyword(s):  
Friction Coefficient ◽  
Kinetic Energy ◽  
Simulation Method ◽  
Road Surface ◽  
Superior Performance ◽  
Time Data ◽  
Braking Performance ◽  
The Road ◽  
Detection Technology ◽  
Road Friction

Anti-lock Braking System (ABS) has widespread used depending on its mature technology and superior performance. We design a test rig which can simulate the running condition of wheels for ABS to detect the braking performance. The kinetic energy of vehicle is replaced by the kinetic energy of rotating flywheel, and the tire-road friction coefficient is replaced by magnetic powder clutch. The amplitude of exciting current to the clutch has linear relationship with the friction coefficient, so as to provide a datum for detecting the working status of ABS under various road conditions. The system can realize simulating test of single road surface, bisectional road surface and joint road surface. The validity of the road simulation method can be verified by the real-time data from the user interface.

scholarly journals Intrinsic motivation and episodic memories for robot exploration of high-dimensional sensory spaces

2020 ◽  
pp. 105971232092291
Author(s):  
Guido Schillaci ◽  
Antonio Pico Villalpando ◽  
Verena V Hafner ◽  
Peter Hanappe ◽  
David Colliaux ◽  
...  
Keyword(s):  
Neural Networks ◽  
Episodic Memory ◽  
Intrinsic Motivation ◽  
Computational Models ◽  
Deep Neural Networks ◽  
Image Sensor ◽  
Forward Kinematics ◽  
High Dimensional ◽  
Episodic Memories ◽  
Low Dimensional

This work presents an architecture that generates curiosity-driven goal-directed exploration behaviours for an image sensor of a microfarming robot. A combination of deep neural networks for offline unsupervised learning of low-dimensional features from images and of online learning of shallow neural networks representing the inverse and forward kinematics of the system have been used. The artificial curiosity system assigns interest values to a set of pre-defined goals and drives the exploration towards those that are expected to maximise the learning progress. We propose the integration of an episodic memory in intrinsic motivation systems to face catastrophic forgetting issues, typically experienced when performing online updates of artificial neural networks. Our results show that adopting an episodic memory system not only prevents the computational models from quickly forgetting knowledge that has been previously acquired but also provides new avenues for modulating the balance between plasticity and stability of the models.

scholarly journals Deep neural networks for the assessment of surgical skills: A systematic review

2021 ◽  
pp. 154851292110345
Author(s):  
Erim Yanik ◽  
Xavier Intes ◽  
Uwe Kruger ◽  
Pingkun Yan ◽  
David Diller ◽  
...  
Keyword(s):  
Systematic Review ◽  
Neural Networks ◽  
Deep Neural Networks ◽  
Surgical Skills ◽  
Video Data ◽  
Skill Assessment ◽  
Assessment Process ◽  
Surgical Skill ◽  
Surgical Trainee ◽  
Surgical Skill Assessment

Surgical training in medical school residency programs has followed the apprenticeship model. The learning and assessment process is inherently subjective and time-consuming. Thus, there is a need for objective methods to assess surgical skills. Here, we use the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to systematically survey the literature on the use of Deep Neural Networks for automated and objective surgical skill assessment, with a focus on kinematic data as putative markers of surgical competency. There is considerable recent interest in deep neural networks (DNNs) due to the availability of powerful algorithms, multiple datasets, some of which are publicly available, as well as efficient computational hardware to train and host them. We have reviewed 530 papers, of which we selected 25 for this systematic review. Based on this review, we concluded that DNNs are potent tools for automated, objective surgical skill assessment using both kinematic and video data. The field would benefit from large, publicly available, annotated datasets representing the surgical trainee and expert demographics and multimodal data beyond kinematics and videos.

scholarly journals METHODS FOR DETERMINING THE FRICTION COEFFICIENT OF THE ROAD SURFACE / TRINTIES KOEFICIENTO SU KELIO DANGA NUSTATYMO METODAI

2019 ◽  
Vol 11 (0) ◽  
pp. 1-6
Author(s):  
Elena Perova ◽  
Evgeniya Ugnenko ◽  
Gintas Viselga ◽  
Ina Tetsman
Keyword(s):  
Friction Coefficient ◽  
Literature Review ◽  
Weather Forecasting ◽  
Weather Conditions ◽  
Air Traffic ◽  
Road Surface ◽  
The Road ◽  
Frictional Properties ◽  
Air Traffic Controllers ◽  
Comprehensive Literature Review

In spite of advances in aviation technology, operational procedures and weather forecasting, safe winter runway operations remain a challenge for airport operators, air traffic controllers, airlines and pilots who must coordinate their actions under rapidly-changing weather conditions. The paper analyses most popular methods for determining the friction coefficient of the road surface. Their advantages, disadvantages and comparison of their modern instruments for measuring the frictional properties of airfields is shown. Most of the information was derived from a comprehensive literature review. Santrauka Nepaisant aviacijos technologijų pažangos, operacinių procedūrų ir oro prognozių, saugios kilimo ir tūpimo tako operacijos žiemą išlieka iššūkiu oro uostų valdytojams, skrydžių vadovams, oro linijoms ir pilotams, kurie turi koordinuoti savo veiksmus sparčiai besikeičiančiomis oro sąlygomis. Straipsnyje analizuojami populiariausieji kelio dangos trinties koeficiento nustatymo metodai. Pateikiami jų privalumai, trūkumai ir lyginami jų modernūs prietaisai, skirti aerodromų trinties savybėms matuoti. Didžioji dalis informacijos buvo gauta iš išsamios literatūros apžvalgos.

scholarly journals On the inference speed and video-compression robustness of DeepLabCut

2018 ◽  
Author(s):  
Alexander Mathis ◽  
Richard Warren
Keyword(s):  
Neural Networks ◽  
Transfer Learning ◽  
Video Compression ◽  
Pose Estimation ◽  
Efficient Method ◽  
Deep Neural Networks ◽  
Video Data ◽  
Frame Size ◽  
Highly Efficient

Pose estimation is crucial for many applications in neuroscience, biomechanics, genetics and beyond. We recently presented a highly efficient method for markerless pose estimation based on transfer learning with deep neural networks called DeepLabCut. Current experiments produce vast amounts of video data, which pose challenges for both storage and analysis. Here we improve the inference speed of DeepLabCut by up to tenfold and benchmark these updates on various CPUs and GPUs. In particular, depending on the frame size, poses can be inferred offline at up to 1200 frames per second (FPS). For instance, 278 × 278 images can be processed at 225 FPS on a GTX 1080 Ti graphics card. Furthermore, we show that DeepLabCut is highly robust to standard video compression (ffmpeg). Compression rates of greater than 1,000 only decrease accuracy by about half a pixel (for 640 × 480 frame size). DeepLabCut’s speed and robustness to compression can save both time and hardware expenses.

scholarly journals Analisa Koefisien Gesek Kinetis Terhadap Struktur Permukaan Jalan Akibat Beban Dinamik Mobil

2018 ◽  
Vol 1 (1) ◽  
pp. 047-051
Author(s):  
Muhammad Nuh Hudawi Pasaribu ◽  
Muhammad Sabri ◽  
Indra Nasution
Keyword(s):  
Friction Coefficient ◽  
Coefficient Of Friction ◽  
Surface Texture ◽  
Road Surface ◽  
Vehicle Speed ◽  
Kinetic Friction ◽  
The Road ◽  
Road Condition ◽  
Asphalt Road ◽  
The Coefficient Of Friction

Tekstur permukaan jalan umumnya terdiri dari aspal dan beton. Kekasaran tekstur permukaan jalan dapat disebabkan oleh struktur perkerasan dan beban kendaraan. Kekasaran tekstur permukaan jalan, bebandan kecepatan kendaraan akan mempengaruhi koefisien gesek. Untuk mengetahui nilai koefisien gesek dilakukan penelitian dengan melakukan variasi beban mobil (Daihatsu Xenia, Toyota Avanza, Toyota Innova dan Toyota Yaris) terhadap kontak permukaan jalan (aspal dan beton) dan kecepatan kendaraan. Hasil penelitian menunjukkan bahwa massa, lebar kontak tapak ban terhadap permukaan jalan dan kecepatan sangat mempengaruhi nilai koefisien gesek kinetis. Koefisien gesek kinetis yang terbesar untuk ketiga kontak permukaan jalan (aspal lama IRI 10,1, Aspal baru IRI 6,4 dan beton IRI 6,7) dengan menggunakan mobil Daihatsu Xenia terjadi pada kondisi jalan beton yaitu 0,495 pada kecepatan 35 Km/Jam. Koefisien kinetis jalan beton > 52 % dibandingkan jalan aspal pada parameter IRI yang sama (6-8).Koefisien gesek kinetis > 0,33 diperoleh di jalan beton pada kecepatan 30 – 40 Km/Jam   The texture of road surface generally consists of asphalt and concrete. The roughness of the road surface texture could be caused by the structure of the pavement and the load of the vehicles. Roughness of road surface texture, load and speed of vehicles would affect to the coefficient of friction. This research was carried out to find out the value of the coefficient of friction by using various load of cars (Daihatsu Xenia, Toyota Avanza, Toyota Innova and Toyota Yaris) on road surface contact (asphalt and concrete) and vehicle speed. The result showed the mass, the width of the tire tread contact to the road surface, and speed very influenced the coefficient value of kinetic friction. The biggest kinetic friction coefficient for all three road surface contacts (IRI 10.1 old asphalt, IRI 6.4 and IRI 6.7) using the Daihatsu Xenia was on the concrete road condition i.e. 0.495 on a speed of 35 km/hour. The concrete road kinetic coefficient was >52% compared to the asphalt road in the same IRI parameter (6-8). The kinetic friction coefficient >0.33 was obtained on the concrete road on a speed of 30 - 40 km/hour.

Sign in / Sign up

Export Citation Format

Share Document