Intersections are quite important and complex traffic scenarios, where the future motion of surrounding vehicles is an indispensable reference factor for decision-making or path planning autonomous vehicles. Considering that trajectory a vehicle at intersection partly obeys statistical law historical data once its driving intention determined, this paper proposes long short-term memory based (LSTM-based) framework combines prediction together. First, we build prior trajectories model (IPTM)...

Intelligent decision making and efficient trajectory planning are closely related in autonomous driving technology, especially highway environment full of dynamic interactive traffic participants. This work integrates them into a unified hierarchical framework with long-term behavior (LTBP) short-term (STDP) running two parallel threads different horizon, consequently forming closed-loop maneuver system that can react to the effectively efficiently. In LTBP, novel voxel structure 'voxel...

Uncontrolled intersections with interaction and uncertainties are challenging for autonomous vehicles (AV) to manage. In this work, we propose a decision-making model specific emphasis on three aspects. First, behavior estimation of the social vehicles’ (SVs) is essential risk avoidance. We try improve prediction accuracy by predicting intentions driving styles SVs in advance doing adaptive goal sampling. Second, uncertainty from results should be considered process. For this, risk-aware...

Trajectory planning for the unmanned vehicle in complex environment has always been a challenging task. Planned trajectory with corresponding target velocity or acceleration sequence must be collision-free guaranteed and as comfortable possible on premise of obeying traffic rules interaction other dynamic social vehicles. To meet this requirement, paper proposes framework based spatio-temporal map. Due to time layer architecture map, can generated simultaneously, whole is constrained within...

As the vital factor of vehicle behavior understanding and prediction, taillight recognition is an important technology for autonomous driving, especially in diverse actual traffic scenes full dynamic interactive participants. However, practical application, it always faces many challenges, such as 'variable lighting conditions', 'non-uniform standards' 'random relative observation pose', which lead to few mature solutions current common autopilot systems. This work proposes action-state...

Uncontrolled intersections are important and challenging traffic scenarios for autonomous vehicles. Vehicles not only need to avoid collisions with dynamic vehicles instantaneously but also predict their behavior then make long-term decisions in reaction. To solve this problem, we propose a cooperative framework composed of Primary Driver (PD) motion planning Subordinate (SD) decision-making. SD is essentially the combination prediction module high-level planner, which develops...

In this paper an improved multi-scale Retinex algorithm is studied to solve the insufficiency that traditional incapable of non-uniform illumination images. Before being processed by algorithm, given images are preprocessed a nonlinear transformation make tends uniform. Besides, combined with PCA and LDA apply for face recognition identify under complex illumination. Simulations show has better ability various illumination, it increases rate

This paper deals with modelling and control method applied to a BAS (Belted Alternator Starter) parallel hybrid electric vehicle. The motor, characterized by high-bandwidth torque control, is here utilized dampen the driveline oscillations that arise during multi-operations, such as tip-in, tip-out, tip-out braking, regeneration, regeneration braking. In order consider new simplified model setup mentioned working operations can be well simulated. By using this model, an MPC-based active...

Prediction is a vital component of motion planning for autonomous vehicles (AVs). By reasoning about the possible behavior other target agents, ego vehicle (EV) can navigate safely, efficiently, and politely. However, most existing work overlooks interdependencies prediction module, only connecting them in sequential pipeline or underexploring results module. In this work, we propose framework that integrates module with three highlights. First, an ego-conditioned model causal prediction,...

This paper investigates the resilient cooperative control of multi-agent systems against potential cyber attacks. The attacker aims to make consensus dynamics diverge by inserting external signals communication network and modify information being exchanged. An event-triggered network-level defense mechanism with virtual is developed, under which system can reach leader-follower in presence unknown strategy proposed this handle attacks without estimation-based compensators has advantage not...

Abstract Driving in interactive dynamic traffic is a huge challenge for autonomous vehicles, especially motion planning. The vehicle not only needs to predict the future states of social vehicles avoid collision but also realizes comfort and continuous plan. To deal with problem, spatio‐temporal decision‐making planning framework flexible constraints based on previous work proposed. Improvements can be highlighted three aspects. First, neural network trajectory prediction trained it...

Considering the characters of information fusion to SINS/GPS, this paper advances a new method random distribution on matrix factors. The simulations show that factors do not effect accuracy estimation when there are no measure errors, but and accelerates convergence. Meanwhile, small quantity calculation needs real-time character.

This paper studies the collective behavior of multi-agent systems with communication delays under influence environment. Leaders in system are able to extract information from environment, while followers not. We extend attraction and repulsion functions a class certain properties, use matrix, graph theory, Lyapunov stability test other methods carry out work. Under general assumption, we prove that individuals will form population gather together enter bounded region. In addition, effect...

In this paper we propose a novel method for image feature extraction. The is developed based on two techniques, i.e., maximum scatter difference (MSD) and diagonal face images projection. addition, an automatic strategy selecting the 2D components or discriminative features adopted. Extensive experiments are performed to test evaluate new algorithm using AR databases. Experimental results show effectiveness of proposed method.

A new method of adaptive binocular vision calibration has been studied, in order to meet the accuracy requirement target tracking mobile robot. On basis theoretical analysis robot, project designed a fuzzy controller. The algorithms controllers have applied robot process complex environment. accuracy, stability and efficiency this controller verified through experiments.

Due to the increasing frequency and intensity of marine monitoring tasks, need for a comprehensive information management system organize implement all types tasks is growing.This paper put forward an integrated using office automation (OA) technology which based on service-oriented architecture (SOA) workflow.The realized in aspects statistical analysis data.And it informationized set processes required.