The over-actuated characteristics of distributed drive electric vehicles (DDEVs) provide a flexible platform to pursue higher holistic performance. This article proposes dual-model predictive control (MPC)-based hierarchical framework realize the energy saving while improving handling stability for DDEVs. upper layer allocates torque vector through front/rear axles, which can high-efficiency zone in-wheel motors and reduce consumption. lower generates direct-yaw-moment (DYC) input by...

Human-machine mutual trust has become one of the important factors restricting steer-by-wire vehicles due to removal mechanical connections. To this end, article proposes a hierarchical shared steering control framework based on human-machine evaluation. The upper level aims evaluate level. driver's in machine is evaluated by difference. And machine's driver skills. lower dynamically optimize authority allocation considering varying states. fuzzy method adopted calculate reference value...

The implementation of direct yaw moment control (DYC) in distributed drive electric vehicles brings greater potential for enhancing vehicle maneuverability and stability performance. Therefore, this paper introduces a robust framework driver assistance by integrating the active front-wheel steering (AFS) system DYC, aiming to improve overall First, behavior is modeled incorporated into path-tracking (DDEV). This integration facilitates comprehensive consideration characteristics during...

Reconstitution of control architecture creates a great challenge for distributed drive electric vehicles (DDEV), due to the emergence new driving strategy. To this end, novel is proposed in paper integrated active front steering (AFS) system and direct yaw moment (DYC) system. First, multi-agent (MAS) employed construct general framework, where AFS DYC act as agents that work together improve vehicle lateral stability simultaneously reduce workloads drivers during path tracking. The...

The uncertainties of driver's behavior seriously affect road safety and bring significant challenges to the human-machine cooperative control. This paper proposes a shared control framework considering time-varying characteristics improve co-driving cooperation performance. Firstly, driving intention is introduced describe involvement level through using Gauss-Bernoulli restricted Boltzmann machine method. And index ability proposed evaluate driver skills based on path-tracking errors. Then,...

Fatigue driving has been regarded as one of the most important factors that cause traffic accidents. This paper proposes a robust human-machine shared control strategy to improve vehicle performance for different driver fatigue states. Firstly, time-varying steering model is proposed address mismatch caused by driving. And evaluation system established based on facial features quantify levels. Based quantified levels, novel allocating authorities and controller developed building...

A human-machine shared steering control is presented in this paper for tracking large-curvature path, considering uncertainties of driver’s characteristics. driver-vehicle-road (DVR) model proposed which uncertain characteristic parameters are defined to describe the human behaviors path. Then radial basis function neural network (RBF) used estimate different drivers’ characteristics and obtain boundaries these parameters. Parameter time-varying vehicle speed DVR handled with Takagi-Sugeno...

Emergency takeover is a common form of human-machine interaction in existing advanced driver assistance systems. It significant to Study the behavioral characteristics drivers during emergency events. However, few studies have considered differences driver's demand for Time-to-Request Lead Time (TORlt) under different risk element compositions and influence experience accumulation on requirements. In this study, 3×3 lane-changing obstacle avoidance simulation experiment was designed with...

This paper presents a modeling method to describe the driver's steering behavior in following trajectory with large curvature. A target scene is established 6-degree-of-freedom driving simulator platform. For collecting experimental data, three experienced drivers are requested drive simulator. Then, back propagating neural network Levenverg-Marquard (LM) algorithm used model large-curvature path. Compared transfer function driver model, proposed BP has higher precision, and potentially more...

Driver distraction behavior is one of the critical factors in traffic accidents. Thus, advanced driver state detection system has become focus field intelligent vehicle. However, practical applications, insufficient samples driving behaviors bring great challenges to training a personalized model for specific driver. To this end, novel transformer based on transfer learning strategy proposed paper accurately recognize behavior. Inspired by effect network visual recognition, we firstly...

Driver drowsiness is an important factor in traffic safety. Thus, many researchers endeavor to develop a reliable driver detection system. However, the large variation of relative face-camera pose and lack large-scale public dataset have brought great challenges train generic model. This paper proposes new network accurately detect from various viewing angles based on transfer learning population-based sampling strategy (TLPSN). Firstly, multitask cascaded convolutional networks (MTCNN)...

Abstract Autonomous driving technology significantly improves road safety, mitigates traffic congestion, and paves the way for a more efficient connected transportation future. However, uncertain scenarios poses great challenge to safe control of autonomous vehicles (AVs). Therefore, this paper proposes an enhanced safety framework by integrating quantified prediction results human behaviors into path-planning process. First, Long Short-Term Memory (LSTM) network is employed driver behavior...

Abstract Under conditions of high automation, drivers’ excessive engagement in non-driving related tasks can severely impact their takeover ability, posing a significant threat to road safety. Therefore, it’s essential explore the channel resources occupied by enhance driving However, many studies have merely categorized different types experiments without clarifying relationship between and they occupy. To determine whether loads under varying impacts on autonomous performance,...

A fuzzy dynamic output-feedback controller is proposed in this paper for assisting driver's steering torque following path of large curvature, considering uncertainties behavior. driver-vehicle-road (DVR) model based on adopted the design. The driver under assumption that utilizes two virtual angles guidance. Seven characteristic parameters are defined to explain behavior, and these considered Takagi-Sugeno (T-S) method handle uncertainties. Then a linear matrix inequalities (LMIs)-based T-S...

This paper proposes a full-order dynamic output feedback shared controller to provide differential drive assisted moment for in-wheel motors (IWMD) electric vehicles. Firstly, preview driver model is applied simulate the human driver's steering behavior. Besides, neuromuscular characteristics are considered in enhancing tactile interaction between and machine. A driver-vehicle coupling behavior vehicle dynamics established. The decomposed into sub-models by considering perturbation of...