Design and implementation of human driving data–based active lane change control for autonomous vehicles (original) (raw)
Related papers
2019
VDM-DM system is developed to address the need to have a comprehensive system that can evaluate the performance of the car and the capability of the driver based on the planned trajectory. This is possible when VDM-DM system integrates the vehicle dynamic response with driver model. Driver model determine the steer input from the geometrical properties of the intended path and this steer angle becomes the input for vehicle dynamic response analysis. Finally, from the position of the car, the steer angle can be calculated. The position of the car will be then compared with the intended path and a new steer input can be determined by the driver model. Two case studies were carried out to demonstrate the application of the VDM-DM in evaluating the performance of the car and the capability of the driver using Double Lane Change (DLC) circuit. Based on the case studies, VDM-DM can be the tool to evaluate the performance of cars and the capability of the drivers. This demonstrates that VD...
Sustainability, 2019
Lane changing of traffic flow is a complicated and significant behavior for traffic safety on the road. Frequent lane changing can cause serious traffic safety issues, particularly on a two-lane road section of a freeway. This study aimed to analyze the effect of significant traffic parameters for traffic safety on lane change frequency using the studied calibrated values for driving logic "conscious" in VISSIM. Video-recorded traffic data were utilized to calibrate the model under specified traffic conditions, and the relationship between observed variables were estimated using simulation plots. The results revealed that changes in average desired speed and traffic volume had a positive relationship with lane change frequency. In addition, lane change frequency was observed to be higher when the speed distribution was set large. 3D surface plots were also developed to show the integrated effect of specified traffic parameters on lane change frequency. Results showed that high average desired speed and large desired speed distribution coupled with high traffic volume increased the lane change frequency tremendously. The study also attempted to develop a regression model to quantify the effect of the observed parameters on lane change frequency. The regression model results showed that desired speed distribution had the highest effect on lane change frequency compared to other traffic parameters. The findings of the current study highlight the most significant traffic parameters that influence the lane change frequency.
Towards a Decision-Making Algorithm for Automatic Lane Change Manoeuvre Considering Traffic Dynamics
PROMET - Traffic&Transportation, 2016
This paper proposes a novel algorithm for decision-making on autonomous lane change manoeuvre in vehicles. The proposed approach defines a number of constraints, based on the vehicle’s dynamics and environmental conditions, which must be satisfied for a safe and comfortable lane change manoeuvre. Inclusion of the lateral position of other vehicles on the road and the tyre-road friction are the main advantages of the proposed algorithm. To develop the lane change manoeuvre decision-making algorithm, first, the equations for the lateral movement of the vehicle in terms of manoeuvre time are produced. Then, the critical manoeuvring time is calculated on the basis of the constraints. Finally, the decision is made on the feasibility of carrying out the manoeuvre by comparing the critical times. Numerous simulations, taking into account the tyre-road friction and vehicles’ inertia and velocity, are conducted to compute thecritical times and a model named TUG-LCA is presented based on the ...
Probabilistic Lane-Change Decision-Making and Planning for Autonomous Heavy Vehicles
IEEE/CAA Journal of Automatica Sinica, 2022
To improve the safety and driving stability of the autonomous heavy truck, it is necessary to consider the differences of driving behavior and drivable trajectories between the heavy trucks and passenger cars. This study proposes a probabilistic decision-making and trajectory planning framework for the autonomous heavy trucks. Firstly, the driving decision process is divided into intention generation and feasibility evaluations, which are realized using the utility theory and risk assessment, respectively. Subsequently the driving decision is made and sent to the trajectory planning module. In order to reflect the greater risks of the truck to other surrounding vehicles, the aggressiveness index (AI) is proposed and quantified to infer the asymmetrical risk level of lane-change maneuver. In the planning stage, the lateral and roll dynamics stability domains are developed as the constraints to exclude the candidate trajectories that would cause vehicle instability. Finally, the simulation results are compared between the proposed model and the artificial potential filed model in the scenarios extracted from the naturalistic driving data. It is shown that the proposed framework can provide the human-like lane-change decisions and truck-friendly trajectories, and performs well in dynamic driving environments.
Modelling Safety Impacts of Automated Driving Systems in Multi-Lane Traffic
2020
The past three decades have witnessed the emergence of several automotive applications that take over the task of vehicle driving on a sustained basis. The most advanced class of such applications is known as Automated Driving Systems (ADSs). ADS can autonomously operate the vehicle on road stretches that fall under its operational design domain. Industry and governments expect that such systems will be technologically feasible shortly and the traffic will be mixed with system-driven and human-driven vehicles. Even though ADSequipped vehicles will have an impact on traffic safety, there is no clarity on if they would enhance or detriment traffic safety and at what conditions and magnitude. A human and an ADS apply fundamentally different processes to acquire information, make decisions, and operate the vehicle. Therefore, our current insights on the relationship between driving behaviour and safety may not be sufficient to predict the possible impacts of ADS systems. Hence there is ...
Safe autonomous lane changes in dense traffic
2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC)
Lane change manoeuvres are complex driving manoeuvres to automate since the vehicle has to anticipate and adapt to intentions of several surrounding vehicles. Selecting a suitable gap to move/merge into the adjacent lane and performing the lane change can be challenging, especially in dense traffic. Existing gap selection methods tend to be either cautious or opportunistic, both of which directly affect the overall availability and safety of the autonomous feature. In this paper we present a method which enables the autonomous vehicles to increase the availability of lane change manoeuvres by reducing the required margins to ensure a safe manoeuvre. The required safety margins are first calculated by making use of the steering and braking capability of the vehicle. It is then shown that this method can be used to perform autonomous lane changes in dense traffic situations with small inter-vehicle gaps. The proposed solution is evaluated by using Model Predictive Control (MPC) to plan and execute the complete motion trajectory.
A Driver Behavior-Based Lane-Changing Model for Urban Arterial Streets
Transportation Science, 2014
Lane-changing algorithms have attracted increased attention during recent years. However, limited research has been conducted to address the probability of changing lanes as a function of driver characteristics and lane-changing scenarios. This study contributes to the development of a comprehensive framework for modeling drivers' lane-changing maneuver on arterials by using driver behavior-related data. Focus group studies and “in-vehicle” driving tests were performed to investigate the effects of driver type under various lane changes on urban arterials and to collect microscopic vehicular data. With these field collected values, a model was developed to estimate the probability of changing lanes under various lane-changing scenarios and to estimate the corresponding gap acceptance characteristics. The lane-changing probability for each scenario was modeled as a function of the factors identified from the focus group discussions, as well as the driver types. In the gap accepta...