Surachet Srinara

Surachet Srinara


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 January 2023 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
SSSurachet Srinara
Surachet Srinara
STSyun Tsai
Syun Tsai
CLCheng-Xian Lin
Cheng-Xian Lin
MTMeng-Lun Tsai
Meng-Lun Tsai
In practice, the accuracy of high-definition-map-based LiDAR localization in normal distribution transform (NDT) scan matching depends on the presence of sufficient appearance information regarding the environment. This paper proposes new strategies for better handling the initialization cope with the initialization and evaluation of LiDAR localization. The first strategy involves establishing a reliable system based on the initial pose and ground truth. This study proposes the integration of an inertial navigation system and a global navigation satellite system. In the second proposed strategy, the point clouds in each partitioned scan area are processed according to the density ratio. The density ratio plays a crucial role in the analysis of NDT pose estimates, which enables a localization system to appropriately determine the uncertainty of the NDT. Preliminary results indicate that the proposed strategies not only can continually provide stable initial poses but also can be used for the reliable evaluation of NDT algorithms. Moreover, geometric information considerably influences the sensitivity on localization accuracy under the presence of sufficient appearance information regarding the environment in NDT pose estimation.

29 July 2024 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
SSSurachet Srinara
Surachet Srinara
YCYu-Ting Chiu
Yu-Ting Chiu
STSyun Tsai
Syun Tsai
MTMeng-Lun Tsai
Meng-Lun Tsai
MAMengchi Ai
Mengchi Ai
High-definition (HD) maps have recently become a key piece of technology in autonomous driving. Over the past few years, various methods and sensors, such as those based on inertial navigation system (INS), global navigation satellite system (GNSS), cameras, and light detection and ranging (LiDAR), have been used to develop HD maps. In this study, we developed novel techniques for enhancing the creation and verification of HD point cloud maps. Firstly, a tightly coupled (TC) INS/GNSS-assisted 3D Normal Distribution Transform (NDT)-LiDAR mapping system has been developed. Utilizing an integrated INS/GNSS, the system provides a reliable initial pose, thereby mitigating the issue of divergence in NDT scan matching, particularly when the vehicle operates at high speeds in challenging LiDAR environments. This approach enhances both navigation accuracy and the precision of the point cloud map. Secondly, alternative ground control points (GCPs) have been established as substitutes for conventional techniques, addressing freeway regulations and managing safety concerns. Thirdly, to ensure the desired accuracy for “where-in-lane” positioning in autonomous vehicle applications, the created point cloud map was validated against the criteria outlined by standardized procedures. Overall, our preliminary results indicate that our HD point cloud map meets the positioning accuracy criteria outlined by the Taiwan Association of Information and Communication Standards. Our point density results also indicate that our generated point cloud map can achieve a high degree of accuracy in in-lane positioning for autonomous vehicle navigation.

10 July 2025 in IEEE Open Journal of Intelligent Transportation Systems
Verified
Journal Article
Open access
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
STSyun Tsai
Syun Tsai
JCJou-An Chen
Jou-An Chen
SSSurachet Srinara
Surachet Srinara
MTMeng-Lun Tsai
Meng-Lun Tsai
NENaser El-Sheimy
Naser El-Sheimy
Efficient parking management is crucial in crowded Asian cities to optimize limited road space and parking facilities. The increasing vehicle ownership rate in Taiwan has intensified the demand for street parking, leading to excessive driving in search of available spots and contributing up to 30% of traffic congestion. This paper proposes a low-cost, infrastructure-free outdoor roadside parking management system based on high-definition (HD) map updating. The system fuses data from a solidstate LiDAR (SSL) system, a monocular camera, an inertial navigation system, a GPS, and HD maps followed by deep-learningbased efficient region extraction. The goal was to achieve high accuracy with minimal computational resources and infrastructure costs. The proposed systems performance for dynamic HD map object updating was evaluated through parking management tests. The systems costs were low due to the selection of SSL and monocular cameras. Traditional and novel extrinsic calibration methods were compared in various experiments, and a hardware architecture for precise sensor time synchronization was designed. Software algorithms for accurate image–point-cloud projection were developed to update HD map parking layers. By using normal distribution transform matching of the SSL and HD point cloud maps, navigation performance was achieved to 0.4-meter accuracy. When applied to license plate localization in two experimental scenarios, the mean performance error was approximately 0.48 and 0.62 m.

26 March 2026 in Sensors
Verified
Journal Article
Open access
PDF file
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
STSyun Tsai
Syun Tsai
CHChi-Hsin Huang
Chi-Hsin Huang
YLYang-En Lu
Yang-En Lu
SSSurachet Srinara
Surachet Srinara
MAMengchi Ai
Mengchi Ai
Autonomous vehicles at level 3 and above must maintain high navigation accuracy, particularly in global navigation satellite system (GNSS)-denied environments. The main innovations of this work are threefold. First, we integrate visual inertial odometry (VIO) and light detection and ranging (LiDAR) inertial odometry (LIO) as external updates to mitigate the rapid drift of micro-electromechanical system (MEMS)-based industrial-grade inertial measurement units (IMUs) during long-term GNSS outages. Second, we adopt a redundant IMU (RIMU) approach that fuses multiple low-cost IMUs to reduce sensor noise and improve reliability. Third, we propose a system calibration methodology using both static and dynamic vehicle motion to estimate extrinsic parameters (boresight angles and lever arms) of the sensors, achieving an overall boresight angle root-mean-square error of 0.04 degrees in the simulation. Experiments were conducted under a 7 min GNSS-denied scenario in an underground parking lot, allowing for comparison of the error characteristics of multi-sensor fusion schemes against a navigation-grade reference. The INS/GNSS/LIO framework achieved a two-dimensional root-mean-square position error of 1.22 m (95% position error within 2.5 m), meeting the lane-level (1.5 m) accuracy requirement under a GNSS outage exceeding 7 min without prior maps. In contrast, the RINS/GNSS/VIO framework yielded a 4.71 m 2D mean position error under the same conditions. This paper provides a quantitative comparison of the baseline error characteristics of VIO-, LIO-, and RIMU-assisted INS/GNSS fusion under a GNSS-denied navigation scenario.

13 December 2023 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Verified
Journal Article
Open access
PDF file
YLY.-E. Lu
Y.-E. Lu
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
MTMeng-Lun Tsai
Meng-Lun Tsai
YCY.-T. Chiu
Y.-T. Chiu
SSSurachet Srinara
Surachet Srinara
Naser El-SheimyNaser El-Sheimy
Naser El-Sheimy
Cost-effective navigation and positioning systems for autonomous vehicles has become a key focus of research in recent years. Having an accurate position within a lane is vital to enabling high levels of automation and improving safety. Traditionally, vehicle navigation and positioning systems have relied heavily on the Global Navigation Satellite System (GNSS), particularly in open-sky scenarios. However, GNSS signals can be easily disrupted by environmental interferences. These include phenomena such as urban canyons, which result from multi-path interferences, as well as challenges posed by Non-Line-of-Sight (NLOS) situations. In the pursuit of developing robust systems resilient to such issues, the concept of sensor fusion has been widely employed. Among all sensors used in commercial self-driving vehicles, mechanical LiDAR is the primary sensor. Utilizing point cloud data from LiDAR and registering it with a prior point cloud map can result in highly accurate position results. However, the high cost of mechanical LiDAR has limited the mass production of point cloud map and autonomous vehicle. In this paper, we evaluate several successful Simultaneous Localization and Mapping (SLAM) architectures from LiDAR-based to LiDAR-Inertial-based using single Solid-State LiDAR (SSL). Last, we proposed a single SSL mapping and localization framework that can achieve 36 centimeters 3D RMSE and 0.5 degree accuracy in heading estimation.

13 December 2023 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Verified
Journal Article
Open access
PDF file
MTMeng-Lun Tsai
Meng-Lun Tsai
Kai-Wei ChiangKai-Wei Chiang
Kai-Wei Chiang
CHC.-Y. Hsieh
C.-Y. Hsieh
SLShixian Lin
Shixian Lin
SSSurachet Srinara
Surachet Srinara
YCY.-T. Chiu
Y.-T. Chiu
In Taiwan, map companies have to referenced the HD Maps guideline and standard which published in TAICS, Taiwan to make ensure the quality of data. The procedure costs lots of times and resources. In order to increase the capability and mileage of HD Maps, this study try to use diverse surveying method and different kinds of sources for ground control points to produce the maps. The test field is National expressway No.8 in Taiwan. With different ground control points sources issue, we measure the features as ground control points and check points from UAS orthophoto. The preliminary results show that the absolute accuracy is suitable the requirement of HD Maps guidelines. With the diverse surveying method topic, we use sensor data from autonomous vehicle and SLAM algorithms to make HD Maps. The preliminary map results can be used as a base map to accomplish the concept of ground control point cloud. It can be used as ground control point to establishment, update, and overlay other point cloud data. On the other hand, these point clouds can be provided for application of SAE Level 3 autonomous vehicle to achieve the positioning accuracy of where in lane (0.5 meter). The flexible producing methods are helpful for formulating and renewing our HD Maps guideline and standard. These experiences will improve the efficiency and integrity for making map.