摘要：【中文】本发明公开了一种融合多传感器数据的轨迹推算方法、介质、终端和装置，将编码器的数据作为机器人位姿状态的预测量，并将IMU的相对水平角度与光流的相对平移量作为观测数据更新机器人的位姿状态，不仅可以有效应对机器人轮子打滑导致的轨迹推算不准确状况，而且可以通过校正机器人的航向提高机器人位姿估计精度，从而适应更多的环境；同时本发明的模型简单实用，保证了机器人轨迹推算的实时性和高精度，具有较强的实际应用意义和商业价值。 【EN】The invention discloses a track calculation method, a medium, a terminal and a device fusing multi-sensor data, which take data of an encoder as a prediction quantity of a robot position and posture state, take relative translation quantity of relative horizontal angle and optical flow of an IMU as observation data to update the position and posture state of the robot, not only can effectively deal with the condition of inaccurate track calculation caused by skidding of wheels of the robot, but also can improve the estimation precision of the robot position and posture by correcting the course of the robot, thereby being suitable for more environments; meanwhile, the model is simple and practical, real-time performance and high precision of robot track calculation are guaranteed, and the method has strong practical application significance and commercial value.

摘要：【中文】本发明公开一种机器人图像仿真数据的构建方法、介质、终端和装置。首先给定了一条在平面上运动的轨迹，然后利用IMU的运动学模型，生成了需要的加速度计值和陀螺仪值，然后给定从摄像头到IMU的变换矩阵，利用IMU的姿态和变换矩阵得到相机的姿态，从而获得投影后的图像，最终得到了在此条运动轨迹下IMU以及对应的图像数据，所生成的这些数据可以作为VIO系统的输入，从而达到测试的目的，提高VIO系统的测试效率和实际运行效果。 【EN】The invention discloses a method, a medium, a terminal and a device for constructing robot image simulation data. Firstly, a motion track on a plane is given, then a required accelerometer value and a required gyroscope value are generated by utilizing a kinematic model of the IMU, then a transformation matrix from a camera to the IMU is given, the attitude of the camera is obtained by utilizing the attitude of the IMU and the transformation matrix, so that a projected image is obtained, finally the IMU and corresponding image data under the motion track are obtained, and the generated data can be used as the input of the VIO system, so that the test purpose is achieved, and the test efficiency and the actual operation effect of the VIO system are improved.

摘要：【中文】本发明公开一种用于人机交互的通信系统和机器人，包括基于安卓系统的处理器，所述处理器通过4G模块和/或WIFI模块连接外网，通过以太网模块连接机器人内部处理器，同时所述安卓系统的中间层网络连接模块用于保持4G模块、WIFI模块和以太网模块同时在线。本发明的通信系统可以取代现有机器人的Linux+Ubuntu或Linux+Debian系统，使主板处理器在选型范围上更广，得到的技术支援和开发资源也更多，从而大幅降低机器人各方面的成本，更重要的是可以将手机方案引入到机器人，特别是配送机器人领域，进一步降低软/硬及开发成本。 【EN】The invention discloses a communication system and a robot for man-machine interaction, which comprise a processor based on an android system, wherein the processor is connected with an external network through a 4G module and/or a WIFI module, and is connected with an internal processor of the robot through an Ethernet module, and meanwhile, a middle-layer network connection module of the android system is used for keeping the 4G module, the WIFI module and the Ethernet module on line at the same time. The communication system of the invention can replace the Linux + Ubuntu or Linux + Debian system of the existing robot, so that the main board processor is wider in the selection range, and the obtained technical support and development resources are more, thereby greatly reducing the cost of the robot in all aspects, more importantly, the mobile phone scheme can be introduced into the robot, especially the field of distribution robots, and further reducing the soft/hard cost and the development cost.

摘要：【中文】本发明公开一种机器人自动重定位方法、介质、终端和装置。方法包括以下步骤：自动判断机器人定位是否丢失，若是，则采用粒子滤波方法获取预设楼宇地图数据中与定位丢失时的传感检测断面特征相匹配的目标地图形状特征，并将所述目标地图形状特征对应的物理位置作为机器人的真实物理位置，将定位系统中机器人的当前位置替换为所述真实物理位置。本发明整个重定位过程无需人工干预，在降低机器人运营成本的同时无需将机器人移动到初始位置或者预设开机点，因此可以继续执行机器人定位丢失前的业务，提高了机器人运营成功率。 【EN】The invention discloses an automatic robot repositioning method, medium, terminal and device. The method comprises the following steps: and automatically judging whether the robot positioning is lost, if so, acquiring a target map shape feature matched with the sensing detection section feature when the positioning is lost in preset building map data by adopting a particle filtering method, taking a physical position corresponding to the target map shape feature as a real physical position of the robot, and replacing the current position of the robot in the positioning system with the real physical position. The whole repositioning process does not need manual intervention, the operation cost of the robot is reduced, and meanwhile, the robot does not need to be moved to an initial position or a preset starting point, so that the service before robot positioning loss can be continuously executed, and the operation success rate of the robot is improved.

摘要：【中文】本发明公开了智能机器人部件寿命全自动测试系统和方法，其中，所述系统包括：传感器采集模块，用于采集机器人上各个部件的实时状态信息；主控模块，用于将获取到的各个部件的实时状态信息上传至PC上位机，并且接受所述PC上位机反馈的控制指令；执行模块，用于根据反馈的控制指令控制所述机器人。克服现有技术中都是对机器人的各个部件分开进行测试，其测试的结果不够真实准确，不能发现实际场景应用下存在的问题。 【EN】The invention discloses a full-automatic testing system and a full-automatic testing method for the service life of intelligent robot parts, wherein the system comprises: the sensor acquisition module is used for acquiring real-time state information of each part on the robot; the main control module is used for uploading the acquired real-time state information of each component to a PC upper computer and receiving a control instruction fed back by the PC upper computer; and the execution module is used for controlling the robot according to the feedback control instruction. The problem that in the prior art, all parts of the robot are separately tested, the test result is not real and accurate enough, and the actual scene application cannot be found is solved.

摘要：【中文】本发明公开了一种多机器人自动停靠方法、介质、终端和装置，包括以下步骤：当机器人到达目标区域时，检测目标区域的当前障碍物信息，并查询目标区域的预设停靠策略表，获取与最近障碍物距离大于预设值的多个停靠位置，并将评分最高的停靠位置作为最佳停靠位置；控制机器人向最佳停靠位置移动，并在移动过程中对最佳停靠位置进行优化。本发明提出了一种基于共识机制的多机器人通过自组织选择对应停靠位置的方法，该方法可以在不依赖机器人之间互相通信，也不依赖中心化调度系统的前提下，使多机器人在同一空间下完成设计好的站位策略并排列出指定队形，同时不会增加系统成本，也不会因通信环境变化出现延迟，适合在多机器人领域广泛应用。 【EN】The invention discloses a method, a medium, a terminal and a device for automatically stopping multiple robots, which comprise the following steps: when the robot reaches a target area, detecting current barrier information of the target area, inquiring a preset parking strategy table of the target area, acquiring a plurality of parking positions with distances to the nearest barrier larger than a preset value, and taking the parking position with the highest score as an optimal parking position; and controlling the robot to move to the optimal parking position, and optimizing the optimal parking position in the moving process. The invention provides a method for selecting corresponding parking positions of multiple robots through self-organization based on a consensus mechanism, which can enable the multiple robots to complete a designed station strategy in the same space and list specified formation in parallel on the premise of not depending on mutual communication among the robots and a centralized scheduling system, and meanwhile, the cost of the system is not increased, delay caused by communication environment change is avoided, and the method is suitable for wide application in the field of the multiple robots.

摘要：【中文】本发明提供一种陀螺仪偏差矫正方法，所述陀螺仪偏差校正方法包括，获取图像采集‑惯性测量固连系统的图像帧数据和惯性测量数据，根据所述图像帧数据获取第一位姿变换数据，根据所述惯性测量数据获取与所述第一位姿变换数据相对应的第二位姿变换数据，根据所述第一位姿变换数据和所述第二位姿变换数据获取陀螺仪偏差。利用本发明，可以结合摄像头和惯性测量单元两者进行位资估算时的优点，使两种传感器在进行位姿估算时形成互补，提高位姿估计的稳定性和准确性。 【EN】The invention provides a gyroscope deviation correction method which comprises the steps of obtaining image frame data and inertial measurement data of an image acquisition-inertial measurement fixed connection system, obtaining first position posture transformation data according to the image frame data, obtaining second position posture transformation data corresponding to the first position posture transformation data according to the inertial measurement data, and obtaining gyroscope deviation according to the first position posture transformation data and the second position posture transformation data. The invention can combine the advantages of the camera and the inertia measurement unit when estimating the position resources, so that the two sensors form complementation when estimating the position and orientation, and the stability and the accuracy of the position and orientation estimation are improved.

摘要：【中文】本发明公开了一种基于激光slam的内存优化方法、介质、终端和装置，方法包括以下步骤：将不同粒子中包含相同地图坐标的多个单元格作为单元格组合，且每个单元格均具有相同的原始共有变量和不同的私有变量；对所有粒子初始化；根据激光束扫描结果改变单元格各个变量的取值，当变量n和变量x，y的取值发生变化时，将变量n和变量x，y由共有变量标记为私有变量；当每个单元格的共有变量均变为私有变量时，释放存储所述原始共有变量的存储空间。本发明在保证运算速度和地图精度的情况下，减少了粒子滤波激光slam构建环境地图时约54.5％‑59％的内存空间，有助于构建大场景地图，具有较强的实际应用意义。 【EN】The invention discloses a memory optimization method, a medium, a terminal and a device based on laser slam, wherein the method comprises the following steps: taking a plurality of cells containing the same map coordinates in different particles as a cell combination, wherein each cell has the same original common variable and different private variables; initializing all particles; changing the values of all variables of the unit cells according to the scanning result of the laser beam, and marking the variable n and the variables x and y as private variables from a common variable when the values of the variable n and the variables x and y are changed; and when the common variable of each cell is changed into a private variable, releasing the storage space for storing the original common variable. Under the condition of ensuring the operation speed and the map precision, the method reduces about 54.5-59% of the memory space when the particle filter laser slam constructs the environment map, is beneficial to constructing a large scene map, and has strong practical application significance.

摘要：【中文】本发明公开了智能仓储管理中心关节机械臂保护方法和保护系统，其中，所述方法包括：获取机械臂当前的位置数据和当前的运动参数；根据所述当前的位置数据和当前的运动参数以控制所述机械臂在安全的位置范围内。该方法克服现有技术中的机械臂在工作过程中的安全性得不到保障，容易发生故障或者失控问题。 【EN】The invention discloses a protection method and a protection system for a joint mechanical arm of an intelligent warehousing management center, wherein the method comprises the following steps: acquiring current position data and current motion parameters of the mechanical arm; and controlling the mechanical arm within a safe position range according to the current position data and the current motion parameters. The method solves the problems that the safety of the mechanical arm in the prior art in the working process cannot be guaranteed, and the mechanical arm is easy to break down or lose control.

摘要：【中文】本发明公开一种激光点云与地图的匹配评价方法、介质、终端和装置，方法包括以下步骤：根据激光点云在当前时刻的先验位姿将激光点云在环境地图上栅格化；计算每个激光点在环境地图上分别对应的最近点集；计算最近点集中所有最近点的均值点，并根据激光点与对应均值点的距离生成激光点的匹配得分；对所有激光点的匹配得分进行求和生成激光点云与环境地图的匹配总分。本发明对于误差较大点具有更强的适应性，能更准确反应出点云与地图匹配的实际效果，不容易陷入局部极小值，提高了匹配精度与容错性，同时在保证运算速度的情况下，能更快地收敛到最优解，在实际应用中具有重要意义。 【EN】The invention discloses a method, a medium, a terminal and a device for evaluating matching of laser point cloud and a map, wherein the method comprises the following steps: rasterizing the laser point cloud on an environment map according to the prior pose of the laser point cloud at the current moment; calculating a closest point set corresponding to each laser point on the environment map; calculating the mean value points of all the closest points in the closest point set, and generating the matching score of the laser point according to the distance between the laser point and the corresponding mean value point; and summing the matching scores of all the laser points to generate a matching total score of the laser point cloud and the environment map. The method has stronger adaptability to points with larger errors, can more accurately reflect the actual effect of matching the point cloud and the map, is not easy to fall into a local minimum value, improves the matching precision and fault tolerance, can more quickly converge to an optimal solution under the condition of ensuring the operation speed, and has important significance in practical application.