欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对移动机器人同时定位与地图创建（SLAM）面对噪声干扰时估计精度低、鲁棒性差的缺点,提出一种基于容积扩展H∞滤波（CEH∞F）的SLAM算法。首先,通过线性误差传播特性将容积变换嵌入到扩展H∞滤波框架中,利用得到的CEH∞F计算SLAM条件转移概率密度,避免雅克比矩阵的计算和线性化误差积累的同时增强了算法的鲁棒性;另外,在每次迭代中更新调节因子γ,将噪声干扰到估计误差最大能量增益控制在较小范围内,进一步增强算法鲁棒性。实验部分将所提算法与扩展卡尔曼滤波SLAM（EKF-SLAM）、无迹卡尔曼滤波SLAM（UKF-SLAM）、容积卡尔曼滤波SLAM（CKF-SLAM）在不同噪声环境下进行了对比。结果表明,CEH∞F-SLAM算法具有良好的稳定性与精度,是一种有效的SLAM算法。
英文摘要：
Aiming at the shortcoming that mobile robot simultaneous localization and mapping（ SLAM） algorithm has the problems of low estimation accuracy and low robustness when facing with noise disturbance,a cubature extended H∞filter based SLAM（ CEH∞F-SLAM） algorithm is proposed. By using statistical linear error propagation method,the cubature transform technique can be embedded into the EH∞F framework,and the obtained estimator named CEH∞F is used to calculate SLAM posterior probability density,which avoids the calculation of Jacobian matrix and linearization error accumulation; meanwhile,the robustness of the algorithm is increased. Besides,the tuning factorγis updated at each iteration,thus the maximum energy gain from noise disturbance to estimated errors is restricted in a small range,so that the robustness of the algorithm is further improved when facing unknown noise disturbance. In experiment,the proposed algorithm is compared with EKF-SLAM,UKF-SLAM and CKF-SLAM algorithms in different noise environment. The results show that the CEH∞F-SLAM algorithm has better stability and accuracy,and is an effective SLAM algorithm.
同期刊论文项目
同项目期刊论文
期刊信息
位置：