智能汽车拟人驾驶风险量化方法研究

驾驶风险量化评估对智能汽车拟人驾驶决策至关重要,针对复杂多任务场景下的驾驶风险量化问题,提出了一种基于 人类风险感知机理的智能汽车驾驶风险量化方法。 首先,利用传感器获取驾驶场景周围环境信息与行驶状态信息,并根据人类 驾驶经验对潜在冲突因素赋值代价,生成驾驶场景代价地图;其次,根据车辆运动状态与拟人驾驶的基本原则,利用高斯函数建 立动态风险模型;最后,结合驾驶场景代价图与动态风险模型实时计算拟人驾驶风险量化值。 仿真结果表明,提出的方法能够 基于人类驾驶经验,计算出动态变化的驾驶风险量化值,应用于智能汽车自动驾驶决策,可产生拟人驾驶行为。     

混响环境下基于倒谱BRIR的双耳互相关声源定位算法

在实际封闭环境中,针对存在混响而导致声源定位性能下降的问题,提出一种基于倒谱双耳房间脉冲响应（Binaural room impulse response,BRIR）的双耳互相关声源定位方法.该方法通过从倒谱BRIR中减去混响分量,然后反变换到时域得到估计的脉冲响应,再与数据库中的头部脉冲响应（Head related impulse response,HRIR）进行互相关运算,最大互相关值相对应的位置就是所估计的声源位置.仿真实验结果表明,提出的算法能减少混响环境中带来的定位误差,提高声源定位的精度.     

加窗插值快速傅里叶变换在滚动轴承故障诊断中的应用

针对滚动轴承故障信号强噪声背景和非线性等特点,为精确识别滚动轴承的故障特征频率,在最小熵解卷积和Teager能量算子解调基础上,提出了一种基于Hanning窗插值快速傅里叶变换的滚动轴承故障诊断新方法。该方法首先利用最小熵解卷积对轴承故障信号进行降噪,再结合Teager 能量算子对降噪后的故障振动信号进行解调,经傅里叶变换后得到信号的Teager解调谱;然后采用Hanning窗对解调谱进行加权处理;最后利用信号频点附近三根离散频谱的幅值做插值处理,从而得到精确的故障特征频率。轴承实测振动信号的分析结果表明：与传统的Teager 能量算子解调方法相比,在选取较少分析点的基础上,大多数情况下所提方法仍能精确识别轴承故障特征频率。     

Finite element analysis on nanoindentation with friction contact at the film/substrate interface

Finite elements (FE) provide a numerical method to calculate complex nanoindentation problems. Correlation of FE analysis with experimental data on nanoindentation may lead to improved characterization of the mechanical properties of thin films and coating systems. In this study, a model of the friction contact at the interface of thin films and substrates is established using FE analysis about a cone indenter, which imitates a Berkovich nanoindenter. Finite element nanoindentation simulations were performed at three different interface friction contact conditions. The following conclusions were depicted through the study of the simulation data. First, for increasing values of the friction coefficient, the indenter’s force versus displacement response of the film/substrate-friction-contact (F/SFC) model coincides with the response of the film/substrate-perfectly-bonded (F/SPB) model. Second, when the indenter’s maximum displacement is less than 10% of the film thickness the deformed nanoindentation area is concentrated under the indenter tip for both F/SFC and F/SPB models. Third, a mechanical response is generated along the F/SFC interface while the mechanical response along the F/SPB interface is negligible. Finally, the nanoindentation simulation data indicate that the calculated mechanical properties intrinsically depend on the interfacial contact conditions of the film/substrate even when the maximum displacement of the indenter is controlled within the 10% of the thin-film thickness.     

A closed-form approach for identification of dynamical contact parameters in spindle–holder–tool assemblies

Accurate identification of contact dynamics is very crucial in predicting the dynamic behavior and chatter stability of spindle–tool assemblies in machining centers. It is well known that the stability lobe diagrams used for predicting regenerative chatter vibrations can be obtained from the tool point frequency response function (FRF) of the system. As previously shown by the authors, contact dynamics at the spindle–holder and holder–tool interfaces as well as the dynamics of bearings affect the tool point FRF considerably. Contact stiffness and damping values alter the frequencies and peak values of dominant vibration modes, respectively. Fast and accurate identification of contact dynamics in spindle–tool assemblies has become an important issue in the recent years. In this paper, a new method for identifying contact dynamics in spindle–holder–tool assemblies from experimental measurements is presented. The elastic receptance coupling equations are employed in a simple manner and closed-form expressions are obtained for the stiffness and damping parameters of the joint of interest. Although this study focuses on the contact dynamics at the spindle–holder and holder–tool interfaces of the assembly, the identification approach proposed in this paper might as well be used for identifying the dynamical parameters of bearings, spindle–holder interface and as well as other critical joints. After presenting the mathematical theory, an analytical case study is given for demonstration of the identification approach. Experimental verification is provided for identification of the dynamical contact parameters at the holder–tool interface of a spindle–holder–tool assembly.     

Analysis of thermal errors in a high-speed micro-milling spindle

Thermally induced errors account for the majority of fabrication accuracy loss in an uncompensated machine tool. This issue is particularly relevant in the micro-machining arena due to the comparable size of thermal errors and the characteristic dimensions of the parts under fabrication. A spindle of a micro-milling machine tool is one of the main sources of thermal errors. Other sources of thermal errors include drive elements like linear motors and bearings, the machining process itself and external thermal influences such as variation in ambient temperature. The basic strategy for alleviating the magnitude of these thermal errors can be achieved by thermal desensitization, control and compensation within the machine tool.This paper describes a spindle growth compensation scheme that aims towards reducing its thermally-induced machining errors. The implementation of this scheme is simple in nature and it can be easily and quickly executed in an industrial environment with minimal investment of manpower and component modifications.Initially a finite element analysis (FEA) is conducted on the spindle assembly. This FEA correlates the temperature rise, due to heating from the spindle bearings and the motor, to the resulting structural deformation. Additionally, the structural deformation of the spindle along with temperature change at its various critical points is experimentally obtained by a system of thermocouples and capacitance gages.The experimental values of the temperature changes and the structural deformation of the spindle qualitatively agree well with the results obtained by FEA. Consequently, a thermal displacement model of the high-speed micro-milling spindle is formulated from the previously obtained experimental results that effectively predict the spindle displacement under varying spindle speeds. The implementation of this model in the machine tool under investigation is expected to reduce its thermally induced spindle displacement by 80%, from 6 microns to less than 1 micron in a randomly generated test with varying spindle speeds.     

基于切削稳定性与表面质量约束的 铣削工艺参数优化研究

铣削工艺系统的动态特性随着刀具夹持长度改变而变化,影响关联的铣削稳定性与加工表面质量,进而导致铣削加工 的工艺规划具有不确定性。 针对此问题,本文综合刀具悬伸量与传统铣削用量作为输入,分别建立极限切削深度与表面粗糙度 的反向传播神经网络模型(BPNN),并进一步以其表达铣削稳定性与加工质量约束,构建以刀具悬伸量和粗/ 精加工阶段铣削 用量为优化变量,以粗/ 精铣总加工时间为目标的铣削工艺参数优化模型,采用麻雀搜索算法( SSA)对模型进行寻优求解。 以 某型数控机床的夹具型腔铣削加工为例,采用刀具悬伸量与各阶段铣削用量的优化配置进行加工实验,总切削时间 12. 577 min 与表面粗糙度值 3. 01 μm 验证了优化模型的可行性和有效性。     

一种多焦距动态立体视觉 SLAM

现有的双目同步定位与建图(SLAM)都使用标准立体相机,所处环境为静态的假设会影响其在动态环境中的精度。 提 出了一种多焦距动态立体视觉 SLAM 方法,它克服了标准立体相机无法兼顾远距离和宽视场感知场景的缺点,并去除了动态物 体对 SLAM 的影响。 具体来说,对传统的立体校正方法进行了改进,并使用校正参数修正了特征点的位置,而不是整张图像,还 提出了一种自适应特征提取和匹配方法以增加多焦距图像的特征匹配数量。 综合使用多视图几何、区域特征流和相对距离检 测动态对象,剔除动态对象上的特征点。 在公开数据集 KITTI 上,该方法相对 ORB-SLAM3 和 DynaSLAM 的定位精度都提高了 6. 97% ,在自建数据集中,该方法的定位精度比 ORB-SLAM3 提高了 26. 64% ,比 DynaSLAM 提高了 32. 09% 。     

Dynamic response of anisotropic sandwich flat panels to explosive pressure pulses

This paper addresses the problem of the dynamic response in bending of flat sandwich panels exposed to time-dependent external blast pulses. The study is carried out in the context of an advanced model of sandwich structures that is characterized by anisotropic laminated face sheets and an orthotropic core layer, and a closed form solution of the dynamic response to a variety of blast pulses is provided. A detailed analysis of the influence of a large number of parameters associated with the particular type of pressure pulses, panel geometry, fiber orientation in the face sheets and, presence of tensile uni/biaxial edge loads on dynamic response is carried out, and pertinent conclusions are outlined.