Search

Virtual traffic simulation plays an important role in easing traffic congestion and reducing traffic pollution. As the transportation network expands, the former rule-based mobility models showed several limitations in producing convincing virtual vehicles. A more realistic model with example-based method is in demand. In this paper, a neural network is employed with carefully selected traffic trajectory data. The virtual vehicle production is driven by the proposed mobility model and organized by a specified structure. Then, the virtual traffic simulation could be given for an indicated scenario. [ABSTRACT FROM AUTHOR]

Designing intelligent and robust autonomous navigation systems remains a great challenge in mobile robotics. Inverse reinforcement learning (IRL) offers an efficient learning technique from expert demonstrations to teach robots how to perform specific tasks without manually specifying the reward function. Most of existing IRL algorithms assume the expert policy to be optimal and deterministic, and are applied to experiments with relatively small-size state spaces. However, in autonomous navigation tasks, the state spaces are frequently large and demonstrations can hardly visit all the states. Meanwhile the expert policy may be non-optimal and stochastic. In this paper, we focus on IRL with large-scale and high-dimensional state spaces by introducing the neural network to generalize the expert’s behaviors to unvisited regions of the state space and an explicit policy representation is easily expressed by neural network, even for the stochastic expert policy. An efficient and convenient algorithm, Neural Inverse Reinforcement Learning (NIRL), is proposed. Experimental results on simulated autonomous navigation tasks show that a mobile robot using our approach can successfully navigate to the target position without colliding with unpredicted obstacles, largely reduce the learning time, and has a good generalization performance on undemonstrated states. Hence prove the robot intelligence of autonomous navigation transplanted from limited demonstrations to completely unknown tasks. [ABSTRACT FROM AUTHOR]

Highlights: [•] A general multi-agent based framework for virtual reality intelligent simulation system of vehicles is proposed. [•] Modeling and formal specifications of multi-agent traffic simulation are studied. [•] Dynamic features of the real world are realized in the simulation system. [•] Microscopic traffic behavior and vehicle control algorithm are verified by the intelligent simulation system. [Copyright &y& Elsevier]

Highlights: [•] Overtaking behavior model based on intelligent vehicles involving environment effect is established. [•] Fuzzy logic based intelligent vehicle controllers adapting to complex environment are designed. [•] Virtual reality combines with Geographic Information System to realize terrain visualization and provide realistic environment information for simulation. [•] Overtaking behavior maneuver considering environment is realized in virtual reality simulation system. [Copyright &y& Elsevier]

The article presents a study which proposes a new hierarchical method for the flexible job-shop scheduling problem (FJSP). The FJSP is a nondeterministic polynomial-time hard since it is an extension of the job-shop scheduling problem. The first method proposed is based successively on a heuristic approach and a local search. The second one is based on a branch-and-bound algorithm.

This paper presents the design of an integrated longitudinal and lateral controller for autonomous vehicle and field tests with an electric vehicle. First, the longitudinal design was studied which includes the spacing policy as the upper level controller and throttle and brake control as the lower level controller. A safety spacing policy was proposed considering both the vehicle states and the vehicle capability. A coordinated throttle and brake controller was also designed to ensure the vehicle pursuing the desired acceleration. Second, a multimodel lateral controller was proposed which can perform the lane tracking and lane changing manoeuvres. Then, an integrated control structure was proposed to manage both the longitudinal and lateral controller. Finally, simulation and visualization works were carried out to validate the proposed solutions. An electric vehicle experiment platform was also built, and field tests showed encouraging results. [ABSTRACT FROM AUTHOR]

Highlights • A cooperative trajectory planning strategy is proposed to pass through an intersection with V2X communication. • A manager center is established to collect and distributes usage occupancy information in conflict zone. • The proposed strategy has the smallest evacuation-time among that in fixed lights and adaptive lights. Abstract This paper proposes a cooperative scheduling mechanism for autonomous vehicles passing through an intersection, called TP-AIM. The main objective of this research is to ensure safe driving while minimizing delay in an intersection without traffic lights. Firstly, an intersection management system, used as an info-collecting-organizing center, assigns reasonable priorities for all present vehicles and hence plans their trajectories. Secondly, a window searching algorithm is performed to find an entering window, which can produce a collision-free trajectory with minimal delay, besides backup windows. Finally, vehicles can arrange their trajectory individually, by applying dynamic programming to compute velocity profile, in order to pass through intersection. MATLAB/Simulink and SUMO based simulations are established among three types of traffic mechanisms with different traffic flows. The results show that the proposed TP-AIM mechanism significantly reduces the average evacuation time and increases throughput by over 20%. Moreover, the paper investigates intersection delay, which can be reduced to less than 10% compared to classical light management systems. Both safety and efficiency can be guaranteed in our proposed mechanism. [ABSTRACT FROM AUTHOR]

Control charts are used to visually identify the signals that define the behavior of industrial processes in univariate cases. However, whenever the statistical quality of more than one critical variable needs to be monitored simultaneously, the procedure becomes much more complicated. This paper presents a methodology on multivariate pattern recognition using the Mahalanobis distance (D 2) and the Support Vector Machine (SVM) technique to recognise two multivariate patterns. The relevance of the study lies in the monitoring of the variables while considering the correlation between them and the effects of interchangeably using a stable multivariate case against an unstable pattern that results in recognition rates up to 91.6 % . [ABSTRACT FROM AUTHOR]