Lateral obstacle avoidance control based on driving behavior recognition of the preceding vehicles in adjacent lanes

2020 ◽  
Vol 68 (10) ◽  
pp. 880-892
Author(s):  
Youguo He ◽  
Xing Gong ◽  
Chaochun Yuan ◽  
Jie Shen ◽  
Yingkui Du
Keyword(s):  
Obstacle Avoidance ◽  
Lateral Displacement ◽  
Longitudinal Velocity ◽  
Driving Behavior ◽  
Lateral Acceleration ◽  
Lateral Motion ◽  
Behavior Recognition ◽  
Dynamic Surface ◽  
Lane Changing ◽  
The Impact

AbstractThis paper proposes a lateral lane change obstacle avoidance constraint control simulation algorithm based on the driving behavior recognition of the preceding vehicles in adjacent lanes. Firstly, the driving behavior of the preceding vehicles is recognized based on the Hidden Markov Model, this research uses longitudinal velocity, lateral displacement and lateral velocity as the optimal observation signals to recognize the driving behaviors including lane-keeping, left-lane-changing or right-lane-changing; Secondly, through the simulation of the dangerous cutting-in behavior of the preceding vehicles in adjacent lanes, this paper calculates the ideal front wheel steering angle according to the designed lateral acceleration in the process of obstacle avoidance, designs the vehicle lateral motion controller by combining the backstepping and Dynamic Surface Control, and the safety boundary of the lateral motion is constrained based on the Barrier Lyapunov Function; Finally, simulation model is built, and the simulation results show that the designed controller has good performance. This active safety technology effectively reduces the impact on the autonomous vehicle safety when the preceding vehicle suddenly cuts into the lane.

Control and Validation of Automated Lane Centering and Lane Changing Maneuver

2009 ◽  
Author(s):  
Jin-Woo Lee ◽  
Bakhtiar Litkouhi
Keyword(s):  
Control Algorithm ◽  
Control Method ◽  
Lateral Displacement ◽  
Lateral Acceleration ◽  
Lane Change ◽  
Time Step ◽  
Lane Changing ◽  
Current Time ◽  
On The Road ◽  
Lateral Offset

This paper describes an automated lane changing control system that has been developed at General Motors Research and Development. This system uses a single monochrome camera to recognize lane markings on the road ahead and uses multiple short range radars to detect surrounding traffic and objects. A sensing unit calculates the host vehicle’s lateral displacement and the heading angle from the center of its lane as well as the relative distance and relative speed of each object. Since the smoothness and comfort of lane change maneuvering are important measures of the control performance, the vehicle dynamic model is integrated into the desired path generation and vehicle’s controller design to reduce the vehicle lateral acceleration and the lateral offset overshoot during the lane change maneuvering. To avoid heavy computation, a simplified model predictive control algorithm is proposed. The control algorithm calculates a steering angle command at the current time step to drive the vehicle to the desired path. The control method and algorithms are implemented on a demonstration vehicle and validated at straight roads and various curve roads of up to 0.001 [1/m] curvature for different vehicle speeds up to 100 [km/hr]. The results show that lane change maneuvering is successfully completed with lateral offset error less than 20 cm.

scholarly journals State Machine Approach for Lane Changing Driving Behavior Recognition

Automation ◽  
2020 ◽  
Vol 1 (1) ◽  
pp. 68-79
Author(s):  
Ruth David ◽  
Sandra Rothe ◽  
Dirk Söffker
Keyword(s):  
Driving Simulator ◽  
Driving Behavior ◽  
State Machine ◽  
The State ◽  
Behavior Recognition ◽  
Lane Changing ◽  
Recognition Model ◽  
Driving Behaviors ◽  
Current State ◽  
Driving Assistance

Research in understanding human behavior is a growing field within the development of Advanced Driving Assistance Systems (ADASs). In this contribution, a state machine approach is proposed to develop a driving behavior recognition model. The state machine approach is a behavior model based on the current state and a given set of inputs. Transitions to different states occur or we remain in the same state producing outputs. The transition between states depends on a set of environmental and driving variables. Based on a heuristic understanding of driving situations modeled as states, as well as one of the related actions modeling the state, using an assumed relation between them as the state machine topology, in this paper, a crisp approach is applied to adapt the model to real behaviors. An important aspect of the contribution is to introduce a trainable state machine-based model to describe drivers’ lane changing behavior. Three driving maneuvers are defined as states. The training of the model is related to the definition/tuning of transition variables (and state definitions). Here, driving data are used as the input for training. The non-dominated sorting genetic algorithm II is used to generate the optimized transition threshold. Comparing the data of actual human driving behaviors collected using driving simulator experiments and the calculated driving behaviors, this approach is able to develop a personalized behavior recognition model. The newly established algorithm presents an easy to apply, reliable, and interpretable AI approach.

Research on behavior recognition based on feature fusion of automatic coder and recurrent neural network

2020 ◽  
Vol 39 (6) ◽  
pp. 8927-8935
Author(s):  
Bing Zheng ◽  
Dawei Yun ◽  
Yan Liang
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Behavior Pattern ◽  
Rapid Development ◽  
Video Data ◽  
Support Vector ◽  
Behavior Recognition ◽  
Learning Methods ◽  
The Impact ◽  
Internet Of Things Technology

Under the impact of COVID-19, research on behavior recognition are highly needed. In this paper, we combine the algorithm of self-adaptive coder and recurrent neural network to realize the research of behavior pattern recognition. At present, most of the research of human behavior recognition is focused on the video data, which is based on the video number. At the same time, due to the complexity of video image data, it is easy to violate personal privacy. With the rapid development of Internet of things technology, it has attracted the attention of a large number of experts and scholars. Researchers have tried to use many machine learning methods, such as random forest, support vector machine and other shallow learning methods, which perform well in the laboratory environment, but there is still a long way to go from practical application. In this paper, a recursive neural network algorithm based on long and short term memory (LSTM) is proposed to realize the recognition of behavior patterns, so as to improve the accuracy of human activity behavior recognition.

scholarly journals Driven Pile Effects on Nearby Cylindrical and Semi-Tapered Pile in Sandy Clay

2021 ◽  
Vol 11 (7) ◽  
pp. 2919
Author(s):  
Massamba Fall ◽  
Zhengguo Gao ◽  
Becaye Cissokho Ndiaye
Keyword(s):  
Coupling Effect ◽  
Lateral Displacement ◽  
Bending Moment ◽  
Adaptive Mesh ◽  
Pile Driving ◽  
Arbitrary Lagrangian Eulerian ◽  
Axial Forces ◽  
Tapered Pile ◽  
Driven Pile ◽  
The Impact

A pile foundation is commonly adopted for transferring superstructure loads into the ground in weaker soil. They diminish the settlement of the infrastructure and augment the soil-bearing capacity. This paper emphases the pile-driving effect on an existing adjacent cylindrical and semi-tapered pile. Driving a three-dimensional pile into the ground is fruitfully accomplished by combining the arbitrary Lagrangian–Eulerian (ALE) adaptive mesh and element deletion methods without adopting any assumptions that would simplify the simulation. Axial forces, bending moment, and lateral displacement were studied in the neighboring already-installed pile. An investigation was made into some factors affecting the forces and bending moment, such as pile spacing and the shape of the already-installed pile (cylindrical, tapered, or semi-tapered). An important response was observed in the impact of the driven pile on the nearby existing one, the bending moment and axial forces were not negligible, and when the pile was loaded, it was recommended to consider the coupling effect. Moreover, the adjacent semi-tapered pile was subjected to less axial and lateral movement than the cylindrical one with the same length and volume for taper angles smaller than 1.0°, and vice versa for taper angles greater than 1.4°.

scholarly journals Integrated, Speckle-Based Displacement Measurement for Lateral Scanning White Light Interferometry

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2486
Author(s):  
Gert Behrends ◽  
Dirk Stöbener ◽  
Andreas Fischer
Keyword(s):  
White Light ◽  
Lateral Displacement ◽  
Displacement Measurement ◽  
Accurate Information ◽  
White Light Interferometry ◽  
Fully Integrated ◽  
Speckle Correlation ◽  
Dsc Measurements ◽  
Scanning White Light Interferometry ◽  
The Impact

Lateral scanning white light interferometry (LSWLI) is a promising technique for high-resolution topography measurements on moving surfaces. To achieve resolutions typically associated with white light interferometry, accurate information on the lateral displacement of the measured surface is essential. Since the uncertainty requirement for a respective displacement measurement is currently not known, Monte Carlo simulations of LSWLI measurements are carried out at first to assess the impact of the displacement uncertainty on the topography measurement. The simulation shows that the uncertainty of the displacement measurement has a larger influence on the total height uncertainty than the uncertainty of the displacing motion itself. Secondly, a sufficiently precise displacement measurement by means of digital speckle correlation (DSC) is proposed that is fully integrated into the field of view of the interferometer. In contrast to externally applied displacement measurement systems, the integrated combination of DSC with LSWLI needs no synchronization and calibration, and it is applicable for translatory as well as rotatory scans. To demonstrate the findings, an LSWLI setup with integrated DSC measurements is realized and tested on a rotating cylindrical object with a surface made of a linear encoder strip.

Parameter Sensitivity Analysis of a Cooperative Dynamic Bus Lane System With Connected Vehicles

2021 ◽  
pp. 036119812110357
Author(s):  
Meng Xie ◽  
Michael Winsor ◽  
Tao Ma ◽  
Andreas Rau ◽  
Fritz Busch ◽  
...  
Keyword(s):  
Travel Time ◽  
Transition Period ◽  
Penetration Rate ◽  
Simulation Models ◽  
Driving Behavior ◽  
Connected Vehicles ◽  
Microscopic Simulation ◽  
Communication Range ◽  
Bus Lane ◽  
The Impact

This paper aims to evaluate the sensitivity of the proposed cooperative dynamic bus lane system with microscopic traffic simulation models. The system creates a flexible bus priority lane that is only activated on demand at an appropriate time with advanced information and communication technologies, which can maximize the use of road space. A decentralized multi-lane cooperative algorithm is developed and implemented in a microscopic simulation environment to coordinate lane changing, gap acceptance, and car-following driving behavior for the connected vehicles (CVs) on the bus lane and the adjacent lanes. The key parameters for the sensitivity study include the penetration rate and communication range of CVs, considering the transition period and gradual uptake of CVs. Multiple scenarios are developed and compared to analyze the impact of key parameters on the system’s performance, such as total saved travel time of all passengers and travel time variation among buses and private vehicles. The microscopic simulation models showed that the cooperative dynamic bus lane system is significantly sensitive to the variations of the penetration rate and the communication range in a congested traffic state. With a CV system and a communication range of 150 m, buses obtain maximum benefits with minimal impacts on private vehicles in the study simulation. The safety concerns induced by cooperative driving behavior are also discussed in this paper.

Understanding the Lateral Dimension of Traffic: Measuring and Modeling Lane Discipline

2021 ◽  
pp. 036119812110318
Author(s):  
Rafael Delpiano
Keyword(s):  
New Technologies ◽  
Autonomous Vehicle ◽  
Longitudinal Velocity ◽  
Lateral Position ◽  
Multiple Cameras ◽  
Lateral Velocity ◽  
Lane Changing ◽  
Lateral Dimension ◽  
Vehicle Position ◽  
Longitudinal Position

There is growing interest in understanding the lateral dimension of traffic. This trend has been motivated by the detection of phenomena unexplained by traditional models and the emergence of new technologies. Previous attempts to address this dimension have focused on lane-changing and non-lane-based traffic. The literature on vehicles keeping their lanes has generally been limited to simple statistics on vehicle position while models assume vehicles stay perfectly centered. Previously the author developed a two-dimensional traffic model aiming to capture such behavior qualitatively. Still pending is a deeper, more accurate comprehension and modeling of the relationships between variables in both axes. The present paper is based on the Next Generation SIMulation (NGSIM) datasets. It was found that lateral position is highly dependent on the longitudinal position, a phenomenon consistent with data capture from multiple cameras. A methodology is proposed to alleviate this problem. It was also discovered that the standard deviation of lateral velocity grows with longitudinal velocity and that the average lateral position varies with longitudinal velocity by up to 8 cm, possibly reflecting greater caution in overtaking. Random walk models were proposed and calibrated to reproduce some of the characteristics measured. It was determined that drivers’ response is much more sensitive to the lateral velocity than to position. These results provide a basis for further advances in understanding the lateral dimension. It is hoped that such comprehension will facilitate the design of autonomous vehicle algorithms that are friendlier to both passengers and the occupants of surrounding vehicles.

scholarly journals The importance of eco-driving training for drivers in transport companies – A case study

2021 ◽  
Vol 67 (1) ◽  
pp. 47-51
Author(s):  
Tatjana Savković ◽  
Pavle Gladović ◽  
Milica Miličić ◽  
Pavle Pitka ◽  
Dejan Koleška
Keyword(s):  
Fuel Consumption ◽  
Management System ◽  
Driving Behavior ◽  
Truck Drivers ◽  
Training Period ◽  
Fleet Management ◽  
Short Term ◽  
The Impact

The paper evaluates the impact of eco-driving programs on driving behavior. The study involved 4 professional truck drivers, which examined two operational driving prameters: fuel consumption and idling. Driving behavior was analyzed through three periods: pre-training period (P1), training period (P2), first month after training (P3) and second month after training (P4). Data were collected using Scania Fleet Management System. The results show that there was an improvement in the observed parameters in short-term. Namely, a decrease in fuel consumption and idling was achieved, in the periods P2, P3 and P4 in relation to the period P1. Due to the realized reductions of the observed parameters, costs in transport companies can be significantly reduced annually.

scholarly journals Investigating the impact of Innovative Treatments on Driving Behavior at School Bus Stops: A Driving Simulator Study

2021 ◽  
Author(s):  
Mustafa Suhail Almallah ◽  
Shabna Sayed Mohammed ◽  
Qinaat Hussain ◽  
Wael K. M. Alhajyaseen
Keyword(s):  
Driving Simulator ◽  
Driving Behavior ◽  
School Bus ◽  
School Buses ◽  
Bus Stop ◽  
Typical Condition ◽  
Simulator Study ◽  
And Control ◽  
The Impact ◽  
Traveling Speed

The illegal overtaking/crossing of stopped school buses has been identified as one of the leading causes of students’ injuries and fatalities. The likelihood of students in getting involved in a school bus-related crash increases during loading/unloading. The main objective of this driving simulator study was to study the effectiveness of different treatments in improving students’ safety by reducing the illegal overtaking/crossing of stopped school buses. Treatments used in this research are LED, Road Narrowing and Red Pavement. All proposed treatments were compared with the control condition (i.e., typical condition in the State of Qatar). Seventy-two subjects with valid Qatari driving license were invited to participate in this study. Each subject was exposed to three situations (i.e., Situation 1: the school bus is stopped in the same traveling direction, Situation 2: the school bus is stopped in the opposite traveling direction, Situation 3: the school bus is not present at the bus stop). Results showed that LED and Road Narrowing treatments were effective in reducing the illegal overtaking/crossing of stopped school buses. Moreover, the stopping behavior for drivers in LED and Road Narrowing was more consistent compared to the Red Pavement and control conditions. Finally, LED and Road Narrowing treatments motivated drivers to reduce their traveling speed by 5.16 km/h and 5.11 km/h, respectively, even with the absence of the school bus. Taking into account the results from this study, we recommend the proposed LED and Road Narrowing as potentially effective treatments to improve students’ safety at school bus stop locations.

The impact of HIV-related neuropsychological dysfunction on driving behavior

2000 ◽  
Vol 6 (7) ◽  
pp. 854-854 ◽  
Author(s):  
THOMAS D. MARCOTTE ◽  
ROBERT K. HEATON ◽  
TANYA WOLFSON ◽  
MICHAEL J. TAYLOR ◽  
OMAR ALHASSOON ◽  
...  
Keyword(s):  
Traumatic Brain Injury ◽  
Positive Predictive Value ◽  
Predictive Value ◽  
Visual Inspection ◽  
Driving Behavior ◽  
Personality Change ◽  
Impaired Consciousness ◽  
Neuropsychological Dysfunction ◽  
Gcs Score ◽  
The Impact

The following is a correction for an error that occurred in the Journal of the International Neuropsychological Society, Vol. 6, No. 3. The error occurred in the article titled “Personality change disorder in children and adolescents following traumatic brain injury,” pp. 279–289, by Max et al. On page 285, under the subheading “Injury Factors,” beginning with the second sentence in the first paragraph, the statement should read:Visual inspection of the distribution of PC relative to lowest post-resuscitation GCS scores revealed that a cut-off of lowest post-resuscitation GCS score of 4 or less versus more than 4 yielded a sensitivity for a diagnosis of persistent PC of 9/14 (64.3%), specificity of 18/23 (78.3%), and a positive predictive value of 0.64 (9.14).A cut-off of duration of impaired consciousness of 100 hr or less versus more than 100 hr yielded a sensitivity for a diagnosis of persistent PC of 11/14 (78.6%), specificity of 20/23 (87.0%), and a positive predictive value (PPV) of 0.79 (11/14).

Sign in / Sign up

Export Citation Format

Share Document