针对煤矿综采工作面设备数量多、位置关系复杂，设备间碰撞检测手段匮乏，预警困难等问题，结合虚拟现实技术(Virtual Reality,VR)、多传感器信息融合技术和包围盒碰撞检测技术，以虚拟环境开发平台为基础，提出一种综采工作面设备远程控制过程中综采工作面的虚拟射线与包围盒相结合的方法，实现了综采工作面设备间碰撞检测和预警监测，详细介绍了基于虚拟现实技术的综采设备碰撞检测和预警监测系统组成与实现方法，以滚筒和护帮板之间的位姿关系为例，对虚拟射线碰撞检测的原理与方法进行了详细说明；进而根据采煤工艺对设备之间的运动姿态进行分析，建立了滚筒-护帮碰撞检测数学模型，高效地实现了采煤机滚筒与液压支架护帮板之间的碰撞检测与预警监测；最后通过搭建的综采设备虚拟远程控制平台进行试验验证，在试验中将虚拟射线检测结果与实际测量结果进行比较，得到其碰撞检测结果的最大误差在49.6 mm之内，满足了实际工况要求。试验表明，虚拟射线检测方法可以高效识别设备间碰撞，并通过设置不同阈值可实现设备间分级碰撞预警监测，对实现综采工作面设备异常检测具有重要意义，可有望解决综采设备间碰撞等异常的实时预警监测。

In view of the problems of large number of equipments,complex position relationship,lack of collision detection means between equipments and difficulty of pre-warning in fully mechanized mining face of coal mine.Combined with Virtual Reality (VR) technology,Multi-sensor Information Fusion (MSIF) technology and Bounding Box Collision Detection (BBCD) technology,this paper puts forward a method of combining virtual ray and bounding box to realize collision detection and early warning monitoring between fully-mechanized mining face equipment in the process of remote control of fully-mechanized mining face equipment,which is based on virtual environment development platform.Firstly,the composition and realization of collision detection and early warning monitoring system for fully-mechanized mining equipment based on virtual reality technology are introduced in detail.Secondly,the principle and method of virtual ray collision detection are described in detail by taking the relationship between the roller of shearer and the guard plate of hydraulic support as an example.Then,according to the coal mining technology,the movement attitude between the equipment is analyzed,and the mathematical model of roller-guard collision detection is established,and the collision detection and pre-warning monitoring between the shearer drum and the hydraulic support guard plate are realized efficiently.Finally,the experiment is verified by the virtual remote control platform of fully-mechanized mining equipment.By comparing the virtual ray detection results with the actual measurement results in the experiment,the maximum error of the collision detection results is within 49.6 mm,which meets the requirements of the actual working condition.The experimental results show that the virtual ray detection method can efficiently identify the collision between devices.It is of great significance to realize the abnormal detection of the fully-mechanized mining face by setting different thresholds to realize the pre-warning and monitoring of the inter-device classification collision.This method will be possible to solve the collision between fully mechanized mining equipment and other abnormal real-time monitoring.

0 引言

1 基于VR的综采设备碰撞检测与预警

2 虚拟射线碰撞检测方法

   2.1 采煤机滚筒目标点位置计算

   2.2 液压支架护帮板目标点位置计算

3 虚拟综采工作面的碰撞检测与预警监测

4 试验及结果分析

5 结论