China
Africa
Explore chapters and articles related to this topic
New Technologies, Vehicle Features, and Technology Development Plan
Published in Vivek D. Bhise, Automotive Product Development, 2017
1.Lane-Departure Warning Systems: A lane-departure warning system provides a warning to a driver when his vehicle begins to move out of its lane (unless a turn signal is activated in that direction of the lane deviation) on freeways and arterial roads (typically while driving over about 40 mph). These systems are designed to minimize run-off-the-road accidents by addressing the main causes of collisions: driver error, distractions, and drowsiness. There are two main types of system: (a) systems that warn the driver (lane-departure warning [LDW]) if the vehicle is leaving its lane by providing visual, audible, and/or vibratory warning (e.g., vibrating the steering wheel), and (b) systems that warn the driver and, if no action is taken, automatically take steps to ensure that the vehicle stays in its lane.
Visual and steering behaviours during lane departures: a longitudinal study of interactions between lane departure warning system, driving task and driving experience
Published in Ergonomics, 2023
Jordan Navarro, Emanuelle Reynaud, Maëlle Pelerin, Marie Claude Ouimet, Catherine Gabaude, Damien Schnebelen
In the A condition, a Lane Departure Warning System (LDWS) was included in the driving simulation as soon as the vehicle started moving. The device was perfectly accurate and delivered an auditory warning for every lane departure. The warning sound was triggered when one of the front wheels of the simulated vehicle reached a lane marking (either on the left or right side of the driving lane). The warning onset was selected based on a series of previous experiments results in order to minimise the risk of false warnings (i.e. too early warnings to be perceived as correct) and maintain a good level of effectiveness (Navarro, Deniel, et al. 2019; Navarro et al. 2016, 2017). The warning sound was defined based on a previous work that showed its effectiveness (Lin et al. 2009). The sound delivered consisted in a 1-second sinusoidal pure tone at 1750 Hz with 6 bursts of 100 ms played at 84 dB, interspersed by 80 ms of silence. The warning sound was played bilaterally until the vehicle returned to a safe position in the driving lane.
Deep learning method for risk identification of autonomous bus operation considering image data augmentation strategies
Published in Traffic Injury Prevention, 2023
The research presented in this article is based on the actual operation data of autonomous bus No. 45 in Shanghai, China. Autonomous bus No. 45 is equipped with assisted driving systems capable of collecting information affecting the safety of the bus’s operation, considering human, vehicle, road, and environmental factors. Firstly, the throttle misstep protection system can collect the accelerator pedal opening data. Secondly, the traction control system can collect data on the speed, revolutions, and GPS direction. Thirdly, the GPS module in the lane departure warning system can collect the latitude and longitude data. Finally, the collision mitigation braking system is able to collect the relative speed, relative lateral distance, and relative longitudinal distance data between the autonomous bus and surrounding vehicles or obstacles, as well as the risk mode information. Research data include the operation data of autonomous bus No. 45 from June 4, 2020, to June 9, 2020. Eight indicators from the data that affect the safety of autonomous bus operation include latitude x_1, longitude x_2, GPS direction x_3, speed x_4, revolution x_5, relative speed x_6, relative lateral distance x_7, and relative longitudinal distance x_8. The risk mode of the autonomous bus can also be obtained from the data, where m_0 means no risk, m_1 means high risk, and m_2 means low risk.
A novel region-based iterative seed method for the detection of multiple lanes
Published in International Journal of Image and Data Fusion, 2020
Suvarna Shirke, Ramanathan Udayakumar
Lane detection is the hot research topic in the field of computer vision and machine learning and has been applied in intelligent vehicle systems (Chuang et al. 2016). Detection of the lane is very important for self-driving vehicles. Lane detection is one of the most significant subsystems for achieving the environmental perception of autonomous vehicles, especially for structured road environment (Fu et al. 2014). The lane detection system is utilised for computing vehicles location and trajectory relative to the lane reliably (Saito et al. 2016). Simultaneously, the lane detection system plays a major role in the lane departure warning system. Nowadays, computer stereo vision has been prevalently utilised for improving the accuracy of the lane detection system (Li et al. 2018). The lane detection system is used in various applications, like LDW, lane bobbing detection, and blind-spot monitoring (Kammel and Pitzer 2008, Borkar et al. 2012). The lane detection task is classified into line and edge detection (Li et al. 2018). In recent years, lane detection is attracted by many researchers, and many works had been implemented for detecting the lane. Beyond providing the vehicle’s location in the lane, multi-lane detection is also used in various applications, like detecting vehicles in multiple lanes and determining lane level accuracy for Global Positioning System (GPS) (Barbari et al. 2006) – based navigation (Jiang et al. 2010).