Links table
Abstract and 1 introduction
1.1. Spatial DVD (SDTS)
1.2. Applications
1.3. Various components of SDTS
1.4. The scope of this work and contributions
2. Related work and 2.1. Digital twins and variables
2.2. Spatial Digital Twin Studies
3. Building blocks of spatial digital twins and 3.1. Get and process data
3.2. Data modeling, storage and management
3.3. Huge data analysis system
3.4. Maps and intermediate programs based on geographic information systems
3.5. Main functional ingredients
4. Other related modern technologies and 4.1. Amnesty International and ML
4.2. Blockchain
4.3. Cloud computing
5. Challenges and future work, and 5.1. Acquire multi -defined and accurate data
5.2. NLP Spatial Information and 5.3. Measurement with databases and huge data platforms for SDT
5.4. Spatial visions and 5.5. Multi -media
5.6. Build an simulator environment
5.7. Imagine complex and varied interactions
5.8. Reducing security and privacy concerns
6. Conclusion and references
5. Challenges and future work
Looking at the current situation of knowledge in spatial techniques and SDTS, we have identified a major set of challenges and opportunities that need immediate attention from researchers and practitioners to build sustainable SDT. In this section, we discuss these challenges and include some important trends to work in the future.
5.1. Acquire multi -defined and accurate data
Most current research [10] In this field, obtaining and completing data is highlighted as one of the main challenges in SDT. SDT includes the acquisition of a wide range of spatial and non -spatial data associated with it. Since SDT needs to use a wide range of devices to capture data of different spatial and temporal accuracy, the quality of this data varies greatly. To our knowledge, no longitudinal study of capturing data has been carried out signs of the bench used to capture different spatial data so that the data is completed from the sources of the difference/devices smoothly. For example, the integration of BIM and 3D GIS data remains a challenge due to the generation process (or data sources) and the differences in the criteria used in these two formats.
5.2. NLP Spatial Information
The current SDTS query technologies are limited to running SQL queries on NOSQL database systems or NOSQL or running text quotes on map services (for example, finding Poi found Google Maps). Modern breakthroughs in NLP enable researchers to set text techniques to SQL that facilitate the automatic translation of the natural language text to SQL and operate the query to recover answers from database schedules[8]. Although the accuracy of such methods is still not good enough for commercial use, the last breakthroughs in the very large speaking language models, such as GPT-3/4 [99] And a new [100]A great promise appears in the treatment of natural language -based query in databases systems. We imagine that there is a large range to research the treatment of the natural language -based query in SDTS by exploiting the strength of these large language models. Since spatial data and relationships, and other associated data that describe spatial entities make the entire data interaction use complex, it will be interesting to see how we can use different spatial properties (for example, adjacent/near) and structural (for example, road network) with table data to answer user ketches on SDTS.
5.3. Measurement with databases and huge data platforms for SDT
Some modern research works evaluate various spatial databases and huge data platforms, but under limited settings. in [43]The authors have compared the performance of the spatial Oracle and Postgresql using a small spatial data collection consisting of New York City blocks and streets, where they used a chosen query and a scale for measuring performance. in [101]The authors compared Mongodb and Postgresql to measure spatial and temporal queries and decision, as they used polygons and the movement of the bowl (sequence of Lat-Long pairs). Another work [102] Comparison of three -based Geospark database technologies, which are Mongodb services, PostGRESQL, and Amachon EC2, where they also used polygons and vessel movement data to measure the performance and proximity. As we have noticed from the current research, the database techniques in this field are still immature and not only support the basic spatial data and spatial information. Since SDT generally hosts different forms of data ranging from 3D construction data to a continuous flow of energy consumption, the effectiveness of delivering this data on current platforms has not yet been determined.
Authors:
(1) Muhammad Yunus Ali, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh;
(2) Mohamed Amer Cheima, College of Information Technology, Monash University, 20 Walk Exhibitions, Clayton, 3164, VIC, Australia;
(3) Tanzima Hashem, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, Ece Building, Dhaka, 1000, Bangladesh;
(4) Anwar Olag, College of Computing, Charles Stort University, Port Makari, 2444, New South Wales, Australia;
(5) Muhammad Ali Babar, College of Computer and Sports Science, Adelaide University, Adelaide, 5005, S, Australia.
