Page 35 - Preliminary Analysis Towards a Standardized Readiness Framework - Interim Report

Preliminary Analysis Towards a Standardized Readiness Framework

(continued)

Examples Potential AI impacts
Characteristics
1. e.g. Speed bumps, Barricades, Banners and Advertisements
2. Route planning
3. Extensions -
4. Fiber to the RSU
5. Computation available on the edge
6. Wireless sensors and capabilities in the vehicle, between the vehi-
cle and RSU, etc
7. In-vehicle Safety accessories (belt, airbags)
Computer and
Deployment capabil- 8. Secure communication networks.
ity: Infrastructure 9. Geo spatial capabilities and infrastructure.
10. energy source: solar panels (for energy autonomy).
11. visualization dashboards and mobile apps
12. Cloud hosting of open data, schemes, policies in machine-read-
able format [49], open portals and real-time updates from
agencies [50]
13. Satellite data coordination [51]
14. Ground stations [52] coordinate the ad-hoc networks for drones
used for disaster management and charging stations for drones.
1. Collision avoidance, Driver attention, Human detection, Local
Innovation, (e.g. Patents, publications, local research)
2. Maturity (e.g. validation, standards compliance, certifications,
labs)
3. AI models (image processing)
4. Real-time In-vehicle measurement of various inertia from the
response times of the driver and signal processing on the driver
controls. Inferred values of DUI levels are output.
5. Estimation algorithms on controls such as fertilizers and pesti-
cides
6. Prediction algorithms on Yield.
7. Classification algorithms on the mapping between crops and
fields. e.g. pest and disease management. Pesticide usage.
Prediction of diseases and irrigation schedules.
Research: Models, 8. Random forest and MARS (Multivariate Adaptive Regression
Algorithms, and Splines) algorithms.
Technology 9. Model (YOLO for unique object counting)
10. CBAM (convolutional block attention mechanism) - model
11. Prediction models such as GWL prediction and Fire Danger
Rating System (FDRS) indices such as Drought Code (DC), Duff
Moisture Code (DMC), and Fire Weather Index (FWI) prediction
based on the GWL. [48]
12. GPT-like static models vs. RAG-based dynamic updates to the
policies database [49].
13. generation (advisory generation) and prediction (forecasting)
14. Fire prediction detection, propagation models
15. RL, multi-agent, collaborative intelligent solution [52]
16. The Models include Khmer ASR (speech engine) , TTS (fast
speech), and chatbot (sentenceBERT for finetuning), for accent
handling more data collection is needed, currently this model is
based on central City. [2]

27