Qinghua Zeng

On the Effect of Localization Errors on Geographic Routing in Sensor Networks

Recently, network localization systems that are based on inter-node ranges have received significant attention. Geographic routing has been considered an application which can utilize the location information from these localization systems. In this paper, we firstly recognize that sensor network localization algorithms generate positioning data with different error patterns compared to those networks where node positions are determined directly from GNSS measurements. Secondly, by simulating practical sensor network scenarios using data from our localization algorithm, we observe that existing geographic routing algorithms in wireless sensor networks (WSNs) adopt very simplistic methods in the treatment of position error, without due consideration of error distribution. Additionally, an insight is given into localization algorithms for WSNs with inhomogeneous error environments. Our observations represent an initial step toward a detailed understanding and design of efficient geographic routing algorithms in location aware WSNs.

Path planning for indoor UAV based on Ant Colony Optimization

UAV autonomous navigation is very useful in many applications, and path planning is one of the key technologies for UAV autonomous navigation. In this paper, the path planning problem to find the optimal path from the start location to the destination in an indoor environment is studied based on Ant Colony Optimization (ACO) algorithm. The workspace of UAV is modeled by applying the grid method, which is usually used in ground robot planning. With the help of 3D gird and new climbing weight parameter, a modified algorithm is presented to solve the premature convergence and low efficiency problem of traditional Ant Colony Optimization algorithm. The simulation results show that these improvements make the search of the optimal path rapidly and efficiently.

Integrating monocular vision and laser point for indoor UAV SLAM

In order to realize the autonomous navigation of an unmanned aerial vehicle while exploring unknown indoor environments and meet the requirements of the indoor UAV positioning and indoor environmental reconstruction, a new strategy is presented based on monocular vision and a laser source. Combining laser-assisted odometry and a scan correlation algorithm with inertial information through an Extended Kaiman Filter (EKF) to realize the purpose localization and mapping. The laser-assisted distance measurement is presented in our paper depends on visual and geometric principle, which has the same effect of laser range finder after error corrections. The pose estimates can be obtained by laser scan matching which is used Sum of Gaussian (SoG) as the essential elements of mapping and localization. At last, in order to improve the autonomy of navigation, EKF is used to fuse inertial information. Experiments in an indoor environment validate the significant engineering reference value of the proposed algorithm.

An Improved Multi-Sensor Fusion Navigation Algorithm Based on the Factor Graph

An integrated navigation system coupled with additional sensors can be used in the Micro Unmanned Aerial Vehicle (MUAV) applications because the multi-sensor information is redundant and complementary, which can markedly improve the system accuracy. How to deal with the information gathered from different sensors efficiently is an important problem. The fact that different sensors provide measurements asynchronously may complicate the processing of these measurements. In addition, the output signals of some sensors appear to have a non-linear character. In order to incorporate these measurements and calculate a navigation solution in real time, the multi-sensor fusion algorithm based on factor graph is proposed. The global optimum solution is factorized according to the chain structure of the factor graph, which allows for a more general form of the conditional probability density. It can convert the fusion matter into connecting factors defined by these measurements to the graph without considering the relationship between the sensor update frequency and the fusion period. An experimental MUAV system has been built and some experiments have been performed to prove the effectiveness of the proposed method.

Dynamic Allan Variance Analysis Method with Time-Variant Window Length Based on Fuzzy Control

To solve the problem that dynamic Allan variance (DAVAR) with fixed length of window cannot meet the identification accuracy requirement of fiber optic gyro (FOG) signal over all time domains, a dynamic Allan variance analysis method with time-variant window length based on fuzzy control is proposed. According to the characteristic of FOG signal, a fuzzy controller with the inputs of the first and second derivatives of FOG signal is designed to estimate the window length of the DAVAR. Then the Allan variances of the signals during the time-variant window are simulated to obtain the DAVAR of the FOG signal to describe the dynamic characteristic of the time-varying FOG signal. Additionally, a performance evaluation index of the algorithm based on radar chart is proposed. Experiment results show that, compared with different fixed window lengths DAVAR methods, the change of FOG signal with time can be identified effectively and the evaluation index of performance can be enhanced by 30% at least by the DAVAR method with time-variant window length based on fuzzy control.

Improved PID control algorithm for quadrotor based on MCS

High precision motion capture system has important significance for the research of the control algorithm of the quadrotor. Focus on the problem of the vibration caused by a large deviation when the traditional PID control appears a large velocity change, an incomplete differential ahead improved PID algorithm is proposed to design the flight control system. In order to improve the accuracy of the control algorithm, the motion capture system is introduced in this paper. The design of the navigation and control system are carried out and the test platform is built. The simulation and experimental results show that the controller has good stability and tracking performance, which lay a good foundation for the design and research of the control algorithm.

Receiver Autonomous Integrity Monitoring for Fixed Ambiguity Precise Point Positioning

There are still many challenges in Precise Point Positioning (PPP) including formulation of mathematical models, fast resolution of integer ambiguities, ambiguity validation and integrity monitoring. Research to date has focused on error modelling and ambiguity resolution. The ambiguity validation and integrity monitoring is still to be investigated in detail. Early research on PPP integrity has addressed the transferability of the Carrier phase based Receiver Autonomous Integrity Monitoring (CRAIM) algorithms developed for conventional Real Time Kinematic positioning (cRTK). However, there are significant differences between cRTK and PPP in the characteristics of the corresponding residual errors. For example, the satellite clock errors are removed in cRTK; while there are still satellites clock errors remaining in PPP after the application of correction products. The magnitude of these residual satellite clock errors depends on the quality of the products used. The residual errors in PPP are expected to be bigger than those in cRTK. These errors have significant negative impacts on ambiguity validation and integrity monitoring. This paper addresses these challenges.A Doubly Non-Central F distribution (DNCF) is justified for the use with popular ratio test for ambiguity validation. The residual errors in the PPP are characterised for the two key processes in RAIM, failure detection and derivation of protection levels. The correction products used for tests were from Centre National d’Etudes Spatiales (CNES). The GNSS measurement data used were from the American National Oceanic and Atmospheric Administration (NOAA). This selection is to ensure that data from same stations used to test the method are not part of the data sets for the generation of correction products. A dataset from 2 NOAA stations was used for testing. Test results show that the PPP algorithm with the DNCF based ambiguity validation can reach sub-decimetre accuracy. The protection levels calculated shown to over-bound the position errors all the time. The relative lower protection levels give the potential for the proposed method to be used in critical high accuracy applications.

Node localization algorithm based on UKF filtering with TOF

Localization of nodes is very important in wireless sensor network. A localization algorithm is proposed based on TOF (time of flight) ranging and Unscented Kalman Filtering (UKF). Firstly, TOF ranging was selected to measure the distance between beacon nodes and unknown nodes. With the TOF data, the distance measurement was obtained and optimized, and an node localization method was achieved by using Unscented Kalman Filtering, which avoided the fault of linearization error in Extended Kalman Filtering (EKF). Simulation results indicate that the algorithm combining TOF Ranging with Unscented Kalman filtering performs well in WSN node localization, and it can improve the positioning accuracy greatly. The algorithm was low-cost, and it can satisfy the application for WSN node localization.

Smartphone Heading Correction Based on Gravity Assisted and Middle Time Simulated-Zero Velocity Update Method

Electronic appliances and ferromagnetic materials can be easily found in any building in urban environment. A steady magnetic environment and a pure value of geomagnetic field for calculating the heading of the smartphone in case of pedestrian walking indoors is hard to obtain. Therefore, an independent inertial heading correction algorithm without involving magnetic field but only making full use of the embedded Micro-Electro-Mechanical System (MEMS) Inertial measurement unit (IMU) device in the smartphone is presented in this paper. Aiming at the strict navigation requirements of pedestrian smartphone positioning, the algorithm focused in this paper consists of Gravity Assisted (GA) and Middle Time Simulated-Zero Velocity Update (MTS-ZUPT) methods. With the help of GA method, the different using-mode of the smartphone can be judged based on the data from the gravity sensor of smartphone. Since there is no zero-velocity status for handheld smartphone, the MTS-ZUPT algorithm is proposed based on the idea of Zero Velocity Update (ZUPT) algorithm. A Kalman Filtering algorithm is used to restrain the heading divergence at the middle moment of two steps. The walking experimental results indicate that the MTS-ZUPT algorithm can effectively restrain the heading error diffusion without the assistance of geomagnetic heading. When the MTS-ZUPT method was integrated with GA method, the smartphone navigation system can autonomously judge the using-mode and compensate the heading errors. The pedestrian positioning accuracy is significantly improved and the walking error is only 1.4% to 2.0% of the walking distance in using-mode experiments of the smartphone.

Real-time drogue detection and template tracking strategy for autonomous aerial refueling

Autonomous aerial refueling technology is an effective solution to extend flight duration of unmanned aerial vehicles, and also a great challenge due to its high risk. A novel real-time drogue detection and tracking strategy of monocular vision system for autonomous aerial refueling is proposed. It uses a direct image registration-based tracking method to ensure reliable and real-time tracking, and an ROI detection based on edge features to solve the tracking drift problem. A multiply patches fusion structure is adopted in the tracking method to improve the tracking accuracy and slow the divergence speed. Finally, various experiments are conducted to validate the proposed image processing strategy. These results show that the proposed strategy obtains a high accuracy as well as the real-time performance, and achieves a better performance than state-of-the-art methods.