Global Navigation Satellite System (GNSS) provides accurate positioning data for vehicular navigation in open outdoor environment. In an indoor environment, Light Detection and Ranging (LIDAR) Simultaneous Localization and Mapping (SLAM) establishes a two-dimensional map and provides positioning data. However, LIDAR can only provide relative positioning data and it cannot directly provide the latitude and longitude of the current position. As a consequence, GNSS/Inertial Navigation System (INS) integrated navigation could be employed in outdoors, while the indoors part makes use of INS/LIDAR integrated navigation and the corresponding switching navigation will make the indoor and outdoor positioning consistent. In addition, when the vehicle enters the garage, the GNSS signal will be blurred for a while and then disappeared. Ambiguous GNSS satellite signals will lead to the continuous distortion or overall drift of the positioning trajectory in the indoor condition. Therefore, an INS/LIDAR seamless integrated navigation algorithm and a switching algorithm based on vehicle navigation system are designed. According to the experimental data, the positioning accuracy of the INS/LIDAR navigation algorithm in the simulated environmental experiment is 50% higher than that of the Dead Reckoning (DR) algorithm. Besides, the switching algorithm developed based on the INS/LIDAR integrated navigation algorithm can achieve 80% success rate in navigation mode switching.

Simultaneous Localization and Mapping (SLAM) in an unknown environment is a crucial part for intelligent mobile robots to achieve high-level navigation and interaction tasks. As one of the typical LiDAR-based SLAM algorithms, the Lidar Odometry and Mapping in Real-time (LOAM) algorithm has shown impressive results. However, LOAM only uses low-level geometric features without considering semantic information. Moreover, the lack of a dynamic object removal strategy limits the algorithm to obtain higher accuracy. To this end, this paper extends the LOAM pipeline by integrating semantic information into the original framework. Specifically, we first propose a two-step dynamic objects filtering strategy. Point-wise semantic labels are then used to improve feature extraction and searching for corresponding points. We evaluate the performance of the proposed method in many challenging scenarios, including highway, country and urban from the KITTI dataset. The results demonstrate that the proposed SLAM system outperforms the state-of-the-art SLAM methods in terms of accuracy and robustness.

Traditional calibration method is usually performed with expensive equipments such as three-axis turntable in a laboratory environment. However in practice, in order to ensure the accuracy and stability of the inertial navigation system (INS), it is usually necessary to recalibrate the inertial measurement unit (IMU) without external equipment in the field. In this paper, a new in-field recalibration method for triaxial accelerometer based on beetle swarm antenna search (BSAS) algorithm is proposed. Firstly, as a new intelligent optimization algorithm, BSAS algorithm and its improvements based on basic beetle antennae search (BAS) algorithm are introduced in detail. Secondly, the nonlinear mathematical model of triaxial accelerometer is established for higher calibration accuracy, and then 24 optimal measurement positions are designed by theoretical analysis. In addition, the calibration procedures are improved according to the characteristics of BSAS algorithm, then 15 calibration parameters in the nonlinear method are optimized by BSAS algorithm. Besides, the results of BSAS algorithm and basic BAS algorithm are compared by simulation, which shows the priority of BSAS algorithm in calibration field. Finally, two experiments demonstrate that the proposed method can achieve high precision in-field calibration without any external equipment, and meet the accuracy requirements of the INS.

Model-free reinforcement learning is a powerful and efficient machine-learning paradigm which has been generally used in the robotic control domain. In the reinforcement learning setting, the value function method learns policies by maximizing the state-action value (Q value), but it suffers from inaccurate Q estimation and results in poor performance in a stochastic environment. To mitigate this issue, we present an approach based on the actor-critic framework, and in the critic branch we modify the manner of estimating Q-value by introducing the advantage function, such as dueling network, which can estimate the action-advantage value. The action-advantage value is independent of state and environment noise, we use it as a fine-tuning factor to the estimated Q value. We refer to this approach as the actor-dueling-critic (ADC) network since the frame is inspired by the dueling network. Furthermore, we redesign the dueling network part in the critic branch to make it adapt to the continuous action space. The method was tested on gym classic control environments and an obstacle avoidance environment, and we design a noise environment to test the training stability. The results indicate the ADC approach is more stable and converges faster than the DDPG method in noise environments.

Based on the 3D Reduced Inertial Sensor System (3D-RISS) and the Machine Learning Enhanced Visual Data (MLEVD), an integrated vehicle navigation system is proposed in this paper. In demanding conditions such as outdoor satellite signal interference and indoor navigation, this work incorporates vehicle smooth navigation. Firstly, a landmark is set up and both of its size and position are accurately measured. Secondly, the image with the landmark information is captured quickly by using the machine learning. Thirdly, the template matching method and the Extended Kalman Filter (EKF) are then used to correct the errors of the Inertial Navigation System (INS), which employs the 3D-RISS to reduce the overall cost and ensuring the vehicular positioning accuracy simultaneously. Finally, both outdoor and indoor experiments are conducted to verify the performance of the 3D-RISS/MLEVD integrated navigation technology. Results reveal that the proposed method can effectively reduce the accumulated error of the INS with time while maintaining the positioning error within a few meters.

Fiber optic gyroscope (FOG) is one of the important components of Inertial Navigation Systems (INS). In order to improve the accuracy of the INS, it is necessary to suppress the random error of the FOG signal. In this paper, a variational mode decomposition (VMD) denoising method based on beetle swarm antenna search (BSAS) algorithm is proposed to reduce the noise in FOG signal. Firstly, the BSAS algorithm is introduced in detail. Then, the permutation entropy of the band-limited intrinsic mode functions (BLIMFs) is taken as the optimization index, and two key parameters of VMD algorithm, including decomposition mode number K and quadratic penalty factor $\alpha$ , are optimized by using the BSAS algorithm. Next, a new method based on Hausdorff distance (HD) between the probability density function (PDF) of all BLIMFs and that of the original signal is proposed in this paper to determine the relevant modes. Finally, the selected BLIMF components are reconstructed to get the denoised signal. In addition, the simulation results show that the proposed scheme is better than the existing schemes in terms of noise reduction performance. Two experiments further demonstrate the priority of the proposed scheme in the FOG noise reduction compared with other schemes.