随机运动目标快速精密激光跟踪关键技术研究

Laser tracking measurement is an important means to obtain spatial coordinate information, which is widely applied in the fields of aeronautics and astronautics, scientific research and advanced manufacturing. During the process of laser tracking measurement, rapid and precise tracking of the cooperative target is the premise to conduct tracking measurement. This project aims at breakthrough in the rapid and precise tracking key technology. Firstly, mechanisms of the influence of photoelectric disturbance on target’s micro displacement detection would be analyzed. Then a novel target displacement detection method combining modulation laser, synchronization FFT with position resolving model is proposed, which can improve inhibition ability to photoelectric interference and ensure target tracking precision. Secondly, to solve the fast and accurate target re-tracking problem after target lost, an infrared optical system with characteristics of large scene depth and large field of view is constructed to obtain the lost target’s imaging information. Under tracking state, the artificial neural network mapping model between the imaging information and accurate spatial position of target is set up in advance. When the target is lost, the accurate position of target will be deduced and re-tracking will be realized immediately by means of the neural network mapping model mentioned above. Finally, an improved current statistics model is built to estimate the target motion state, and then to improve the fast response ability of tracking system; on the basis of universe-fuzzy theory, an intelligent self-adaptive control strategy using variable universe-fuzzy algorithm is proposed to improve adaptability for different target states and motion properties. The above research has great significance in theory and application to improve spatial information capturing ability and develop high-performance laser tracking measurement system independently.

激光跟踪测量是空间信息获取的重要手段，在航空航天、科学研究和先进制造等领域具有重要应用。在激光跟踪测量过程中，对目标的快速精密跟踪是实现跟踪测量的前提。本项目以突破激光跟踪关键技术为目标，首先研究光电扰动对目标偏差探测的影响机理，提出将调制激光、同步FFT和位置解算模型相结合的目标偏差探测新方法，提高系统对光电干扰与冲击的抑制能力，保证系统跟踪精度；其次针对目标丢失后的跟踪恢复，采用红外大景深大视场主动光学系统，获取丢失目标的成像信息，通过跟踪状态下预先建立的目标成像信息与目标空间位置之间的神经网络映射模型，实现丢失目标快速跟踪恢复；最后建立改进的“当前”统计模型，用于目标运动状态精确估计，提高跟踪系统的快速反应能力，采用变论域模糊理论研究智能自适应控制算法，提高跟踪系统对目标不同状态和运动特性的适应能力。本项目对提高空间信息获取能力、自主研制高性能激光跟踪测量系统具有重要理论意义和应用。

我国航空航天、军事国防、汽车船舶、轨道交通等高端装备制造以及核能、天文等大科学工程都对激光跟踪测量技术提出了迫切需求，而快速精密激光跟踪控制是实现激光跟踪测量的前提。.本项目以突破快速精密激光跟踪关键技术为目标，重点在六个方面开展了深入研究：.（1）微小目标跟踪偏差精密探测方法：研究了光电扰动对目标偏差探测的影响机理，推导建立了新的位置解算模型，建立了抗干扰数字滤波算法，使激光光斑位置测量稳定性优于1μm，提高了系统对光电干扰与冲击的抑制能力，保证了系统跟踪精度；.（2）大视场目标主动探测方法：提出并构建了红外大景深大视场主动光学系，分析了探测目标（靶球）的主要特征，研究了目标识别策略与靶球中心定位方法，探索了基于深度学习技术的复杂背景下目标的智能检测方法，获得了准确的特征信息，识别准确率高于98%；.（3）激光跟踪初始参数标定方法：研究提出了激光跟踪零点、像旋、比例参数的标定方法，构建了专用标定装置，提高了标定速度，保证了跟踪精度。.（4）目标运动状态估计方法：基于目标的距离、角度、脱靶量等信息，建立了改进的“当前统计”模型，使得估计方法更为切合实际，并基于卡尔曼滤波方法实现了目标的运动状态的在线准确估计；.（5）复合跟踪控制策略：推导了激光跟踪系统的型别，证明了引入前馈环节构成复合控制，才能使稳态误差为0进而实现高性能激光跟踪控制，设计了基于多模PID的复合控制结构，提高了系统对目标跟踪的主动适应性；.（6）试验验证：构建了综合验证平台，开发了激光跟踪专用软件，开展了大范围跟踪、快速机动跟踪、激光跟踪恢复等系列化的试验，实现最远跟踪距离大于71m，跟踪加速度大于1g，跟踪恢复成功率优于98%，这些结果验证了所提出方法的有效性。.本项目系统研究并发展了激光跟踪控制相关理论方法和新技术，突破了快速精密激光跟踪控制技术，推动了激光测量技术的快速发展，提升了激光跟踪测量系统的整体技术水平，应用前景广阔。