In this paper, we demonstrate a system for temporally scalable visual SLAM using a reduced pose graph representation. Unlike previous visual SLAM approaches that maintain static keyframes, our approach uses new measurements to continually improve the map, yet achieves efficiency by avoiding adding redundant frames and not using marginalization to reduce the graph. To evaluate our approach, we present results using an online binocular visual SLAM system that uses place recognition for both robustness and multi-session operation. Additionally, to enable large-scale indoor mapping, our system automatically detects elevator rides based on accelerometer data. We demonstrate long-term mapping in a large multi-floor building, using approximately nine hours of data collected over the course of six months. Our results illustrate the capability of our visual SLAM system to map a large are over extended period of time.

Temporally scalable visual SLAM using a reduced pose graph

Whether Henry Ford or someone else gave us this famous quote, ‘If I had asked people what they wanted, they would have said faster horses’, we may agree that it implies there is a limit to what we can expect from the performance of an existing solution. Science and technology always try to push the boundaries and ‘improve’; improving the quality of our lives or improving the quality of technologies. We, as researchers in the area of navigation, are no exception; we want to improve the quality of navigation services. And there are many ways to do so, and challenges and limitations to those attempts. Some researchers look to improve the accuracy, the reliability, the integrity through different approaches. Some try to reduce or model noise, some try to minimise human error, and some use novel techniques and algorithms for better prediction. Of course, when ‘our horses cannot go any faster’ and there is not much space for improvement for a certain technology or service, researchers may come up with a completely new solution, such as an automobile. Almost all new technologies go through the same exploration period; at the beginning, we want to see how and if it works so we try simple tasks, but then we become more ambitious (or greedier!) and so we introduce it to more difficult challenges until it hits the breaking point. At this point, curious researchers and inventors try to push the boundaries and make the technology better, and if improvement is not possible, they build (invent) a new solution. But what is the ‘quality’ that many of us want to improve? How the quality of a technology or service can be measured in the first place?

/pdf/how-fast-can-our-horses-go-measuring-the-quality-of-1dixa3vt6a.pdf

How Fast Can Our Horses Go? Measuring the Quality of Positioning Technologies

To solve the problem that the IMU nodes of the traditional factor graph are unable to be corrected in time and the positioning accuracy decreases in the condition of satellite failure, an improved preintegration method for the vehicle navigation system is proposed. Different from the current factor graph optimization method, the proposed method that involves corrective parameters of visual sensors is introduced. The structure of the system is changed when GNSS is unavailable and the calculated information of Lidar is used to compensate for IMU errors. The sliding window is used to improve the real-time performance of the system. The experimental results show that the proposed algorithm can improve the positioning accuracy of vehicle navigation when GNSS is unavailable. Compared with the traditional factor graph integrated navigation algorithm, the positioning accuracy of the proposed preintegration method is improved by more than 16% in GNSS failure environments. 

Improved Preintegration Method for Vehicle Navigation Based on Factor Graph Optimization

The superior performance of factor graphs compared to Kalman filtering in various fields and the use of factor graph algorithms instead of Kalman filtering algorithms in moving target localization tasks can reduce target localization error by more than 50%. However, the global factor graph algorithm may cause computational delays due to excessive computational effort. A moving target localization algorithm based on a combination of global and incremental optimization with improved factor graphs is proposed to improve localization accuracy and ensure that the computation time can be adapted to the requirements of online location. A reference point is introduced into the incremental calculation process, and it is first determined whether global or incremental calculation is used for this calculation by comparing the distance between the incremental localization results of the calculated reference point. The position of the UAV itself is then corrected by determining the position of the reference point, and this is used to finally locate the target. Simulation results show that the algorithm has good real-time performance compared to the time-consuming global algorithm. The online localization error of moving targets can be reduced by 17% compared to the incremental calculation results of the classical factor graph algorithm.

/pdf/a-moving-target-online-location-algorithm-based-on-factor-16899f8k.pdf

A Moving Target Online Location Algorithm Based on Factor Graph

To solve the problem that the IMU nodes of the traditional factor graph are unable to be corrected in time and the positioning accuracy decreases in the condition of satellite failure, an improved preintegration method for the vehicle navigation system is proposed. Different from the current factor graph optimization method, the proposed method that involves corrective parameters of visual sensors is introduced. The structure of the system is changed when GNSS is unavailable and the calculated information of Lidar is used to compensate for IMU errors. The sliding window is used to improve the real-time performance of the system. The experimental results show that the proposed algorithm can improve the positioning accuracy of vehicle navigation when GNSS is unavailable. Compared with the traditional factor graph integrated navigation algorithm, the positioning accuracy of the proposed preintegration method is improved by more than 16% in GNSS failure environments.

Visual–Inertial Navigation Systems (VINS) plays an important role in many navigation applications. In order to improve the performance of VINS, a new visual/inertial integrated navigation method, named Sliding-Window Factor Graph optimised algorithm with Dynamic prior information (DSWFG), is proposed. To bound computational complexity, the algorithm limits the scale of data operations through sliding windows, and constructs the states to be optimised in the window with factor graph; at the same time, the prior information for sliding windows is set dynamically to maintain interframe constraints and ensure the accuracy of the state estimation after optimisation. First, the dynamic model of vehicle and the observation equation of VINS are introduced. Next, as a contrast, an Invariant Extended Kalman Filter (InEKF) is constructed. Then, the DSWFG algorithm is described in detail. Finally, based on the test data, the comparison experiments of Extended Kalman Filter (EKF), InEKF and DSWFG algorithms in different motion scenes are presented. The results show that the new method can achieve superior accuracy and stability in almost all motion scenes.

https://www.cambridge.org/core/services/aop-cambridge-core/content/view/523C7C41D18A8D7C159C59235DF502D0/S0373463320000582a.pdf/div-class-title-a-new-visual-inertial-integrated-navigation-algorithm-based-on-sliding-window-factor-graph-optimisation-div.pdf

Feng Jianxin

Papers

A new visual/inertial integrated navigation algorithm based on sliding-window factor graph optimisation