Real-Time Performance-Focused Localization Techniques for Autonomous Vehicle: A Review

Abstract

Real-time, accurate, and robust localisation is critical for autonomous vehicles (AVs) to achieve safe, efficient driving, whilst real-time performance is essential for AVs to achieve their current position in time for decision making. To date, no review paper has quantitatively compared the real-time performance between different localisation techniques based on various hardware platforms and programming languages and analysed the relations among localisation methodologies, real-time performance and accuracy. Therefore, this paper discusses the state-of-the-art localisation techniques and analyses their overall performance in AV application. For further analysis, this paper firstly proposes a localisation algorithm operations capability (LAOC)-based equivalent comparison method to compare the relative computational complexity of different localisation techniques; then, it comprehensively discusses the relations among methodologies, computational complexity, and accuracy. Analysis results show that the computational complexity of localisation approaches differs by a maximum of about times, whilst accuracy varies by about 100 times. Vision-and data fusion-based localisation techniques have about 2-5 times potential for improving accuracy compared with lidar-based localisation. Lidar-and vision-based localisation can reduce computational complexity by improving image registration method efficiency. Data fusion-based localisation can achieve better real-time performance compared with lidar-and vision-based localisation because each standalone sensor does not need to develop a complex algorithm to achieve its best localisation potential. Vehicle-toeverything (V2X) technology can improve positioning robustness. Finally, the potential solutions and future orientations of AVs' localisation based on the quantitative comparison results are discussed.

Human injury-based safety decision of automated vehicles

“…We assumed that time delays with the S3-IRM algorithm are marginal and almost negligible in the future because of the availability of V2VC and cloud computing platforms. The above parameters regarding errors and time delays were determined according to the state-of-the-art technologies ( Table S6 ) ( Lu et al., 2021 ; Yeong et al., 2021 ; Lin et al., 2019 ; Choi et al., 2018 ).…”

Localization for Autonomous Vehicles in Underground Mines

“…Better convergence rate than the sum-mixture, but with lower risk of getting stuck in local minima than for the max-mixture. In [91] ambiguities in observations are modeled as components in a mixture. The max-mixture method is utilized but with a consensus driven possibility to dynamically re-initialize a landmark value to allow the solution to escape a local minima.…”

Roadside Units-Enabled Cooperative Localization Scheme for Autonomous Vehicles Under Gnss-Denied Environments: Methodology and Demonstration

“…The Society of Automotive Engineers (SAE) defines six levels of automation for AVs, ranging from 0 (fully manual) to 5 (fully autonomous) [4]. To achieve a higher level of automation, it is crucial to ensure real-time, accurate and robust vehicle positioning by perceiving the complex driving environment [5]. Conventionally, the positioning methods used for land vehicles employ an integrated system that combines a global navigation satellite system (GNSS) and an inertial navigation system (INS) [6].…”

“…However, it requires a map database and cannot guarantee lane-level accuracy while GNSS signals are unavailable. Another solution is simultaneous localization and mapping (SLAM), which builds an environment model while locating vehicles [5]. SLAM algorithms focus on abstract data from on-board perception sensors, such as Lidar [10], Radar [9], Camera [11], or their combination [15].…”

“…The V2X signal-based positioning method can achieve global positioning of vehicles when GNSS signals are weak or unavailable. Additionally, it can solve non-line-of-sight (NLOS) issues related to on-board perception sensors at a relatively low cost [5], [17], [23]. The fundamental principle of this localization approach involves measuring physical quantities of V2X channel state information, such as time of arrival (TOA), time difference of arrival (TDOA), angle of arrival (AOA), and received signal strength (RSS) [23]- [26].…”