Robust Panoramic SLAM Tightly Coupled With IMU And Wheel Encoder

In this paper,a panoramic simultaneous localization and mapping(SLAM)system tightly coupled with IMU and wheel encoder is proposed to achieve high robust positioning,which can be used in the ground mobile mapping system(MMS),ground mobile robot and driverless car.Due to the natural complementarity of IMU and vision,the research of visual inertial odometry(VIO)becomes very important,and the combination of panoramic vision can play the advantages of vision more effectively.Generally,VIO has four unobservable dimensions of three translations and one yaw angle.However,once the carrier of the VIO system is a ground moving platform,it will cause additional scale unobservability due to uniform acceleration motion or additional pitch angle and roll angle unobservability due to non rotation motion.The introduction of wheel speed sensor can effectively solve this problem.The slam system proposed in this paper consists of the following four parts: 1)measurement data preprocessing;2)system initialization;3)tracking;4)local mapping.The core of the system is the combined adjustment we proposed for the observations of the panoramic camera,IMU,and wheel encoder,to optimize the system state,which includes the rotation,translation,velocity,IMU biases,and local map point coordinates in the initialization,tracking,and local mapping modules to achieve high precision positioning,and to realize high robustness positioning in complex scenes by multisensor fusion and complementarity.In the measurement data preprocessing module,this paper deduces the preintegration form of wheel encoder based on the two wheel differential model,and combines the measurement values of wheel encoder between two frames into a relative motion constraint.The system initialization consists of two steps: 1)the first step is to initialize the visual scale and parameters of the IMU;2)the second step is to perform the combined adjustment for the camera,IMU and preintegrated wheel encoder data to refine the initial parameters.The proposed panoramic camera-IMU-wheel SLAM(PIW-SLAM)system can achieve highprecision and robust localization in challenging scenes through multi-sensor fusion.We validated the performance of the proposed system on seven different challenging data sequences from the University of Michigan North Campus Long-Term Vision and Li DAR Dataset(NCLT)dataset.The results showed that the PIW-SLAM system has a higher localization accuracy compared to the other state-of-the-art SLAM systems,and it also showed a superior robustness in various complex environments.We also verified the accuracy of the scale obtained by the initialization algorithm.Furthermore,we confirmed the performance of the proposed PIW-SLAM system in a wide-baseline scene of the ground motion measurement system by artificially sampling visual frames.