added flight assessment agent
This commit is contained in:
OwusuBlessing
@@ -1,329 +0,0 @@
|
||||
import asyncio
|
||||
import aiohttp
|
||||
import json
|
||||
from datetime import datetime, timedelta
|
||||
from typing import Dict, List, Optional, Tuple
|
||||
import re
|
||||
|
||||
class DroneWeatherDataExtractor:
|
||||
"""
|
||||
Extract location data from drone booking forms and fetch structured
|
||||
weather information for later use in analysis systems.
|
||||
"""
|
||||
|
||||
def __init__(self):
|
||||
self.base_url = "https://api.open-meteo.com/v1/forecast"
|
||||
|
||||
def extract_coordinates(self, booking_data: Dict) -> Tuple[float, float]:
|
||||
"""Extract latitude and longitude from booking form data."""
|
||||
try:
|
||||
# Try to get from GPS coordinates field
|
||||
gps_coords = booking_data.get("site_information", {}).get("gps_coordinates", {})
|
||||
|
||||
if isinstance(gps_coords, dict):
|
||||
lat_str = gps_coords.get("latitude", "")
|
||||
lng_str = gps_coords.get("longitude", "")
|
||||
|
||||
# Parse coordinate strings like "53.4408° N" and "2.2426° W"
|
||||
lat_match = re.search(r'([\d.]+)', lat_str)
|
||||
lng_match = re.search(r'([\d.]+)', lng_str)
|
||||
|
||||
if lat_match and lng_match:
|
||||
latitude = float(lat_match.group(1))
|
||||
longitude = float(lng_match.group(1))
|
||||
|
||||
# Handle direction indicators
|
||||
if 'S' in lat_str.upper():
|
||||
latitude = -latitude
|
||||
if 'W' in lng_str.upper():
|
||||
longitude = -longitude
|
||||
|
||||
return latitude, longitude
|
||||
|
||||
# Fallback: try to extract from site_location string
|
||||
site_location = booking_data.get("form", {}).get("site_location", "")
|
||||
coord_pattern = r'GPS: ([\d.]+)° ([NS]), ([\d.]+)° ([EW])'
|
||||
match = re.search(coord_pattern, site_location)
|
||||
|
||||
if match:
|
||||
lat, lat_dir, lng, lng_dir = match.groups()
|
||||
latitude = float(lat) if lat_dir == 'N' else -float(lat)
|
||||
longitude = float(lng) if lng_dir == 'E' else -float(lng)
|
||||
print(f"Extracted coordinates: {latitude}, {longitude}")
|
||||
return latitude, longitude
|
||||
|
||||
except (ValueError, KeyError) as e:
|
||||
print(f"Error extracting coordinates: {e}")
|
||||
|
||||
raise ValueError("Could not extract valid coordinates from booking data")
|
||||
|
||||
def extract_booking_info(self, booking_data: Dict) -> Dict:
|
||||
"""Extract relevant booking information in structured format."""
|
||||
try:
|
||||
# Extract date information
|
||||
assigned_date = booking_data.get("job_overview", {}).get("assigned_date")
|
||||
preferred_dates = booking_data.get("form", {}).get("preferred_dates")
|
||||
selected_slot = booking_data.get("form", {}).get("selected_slot")
|
||||
|
||||
# Extract timing information
|
||||
timing = booking_data.get("timing", {})
|
||||
start_time = timing.get("start_time")
|
||||
end_time = timing.get("end_time")
|
||||
duration = timing.get("survey_duration")
|
||||
|
||||
# Extract site information
|
||||
site_info = booking_data.get("site_information", {})
|
||||
form_info = booking_data.get("form", {})
|
||||
|
||||
return {
|
||||
"job_id": booking_data.get("job_id"),
|
||||
"job_number": booking_data.get("job_overview", {}).get("job_number"),
|
||||
"site_name": site_info.get("site_name"),
|
||||
"region": site_info.get("region"),
|
||||
"full_address": site_info.get("full_address"),
|
||||
"asset_type": form_info.get("asset_type"),
|
||||
"system_size": form_info.get("system_size"),
|
||||
"survey_purpose": form_info.get("survey_purpose"),
|
||||
"assigned_engineer": form_info.get("assigned_engineer"),
|
||||
"contact_person": form_info.get("contact_person"),
|
||||
"contact_phone": form_info.get("contact_phone"),
|
||||
"dates": {
|
||||
"assigned_date": assigned_date,
|
||||
"preferred_dates": preferred_dates,
|
||||
"selected_slot": selected_slot
|
||||
},
|
||||
"timing": {
|
||||
"start_time": start_time,
|
||||
"end_time": end_time,
|
||||
"duration": duration,
|
||||
"buffer_time": timing.get("buffer_time")
|
||||
},
|
||||
"access_details": {
|
||||
"access_type": form_info.get("access_type"),
|
||||
"vehicle_access": form_info.get("vehicle_access"),
|
||||
"access_details": form_info.get("access_details")
|
||||
},
|
||||
"safety_requirements": form_info.get("special_safety_requirements"),
|
||||
"additional_requirements": form_info.get("additional_requirements"),
|
||||
"booking_timestamp": form_info.get("booking_timestamp"),
|
||||
"booking_id": form_info.get("booking_id")
|
||||
}
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error extracting booking info: {e}")
|
||||
return {}
|
||||
|
||||
def parse_target_date(self, booking_data: Dict) -> datetime:
|
||||
"""Extract and parse the target date from booking data."""
|
||||
try:
|
||||
# Try different date fields
|
||||
date_fields = [
|
||||
booking_data.get("job_overview", {}).get("assigned_date"),
|
||||
booking_data.get("form", {}).get("preferred_dates"),
|
||||
booking_data.get("form", {}).get("selected_slot")
|
||||
]
|
||||
|
||||
for date_field in date_fields:
|
||||
if date_field:
|
||||
# Handle different date formats
|
||||
if "January 23, 2025" in date_field or "January 23rd, 2025" in date_field:
|
||||
return datetime(2025, 1, 23)
|
||||
elif "January 24" in date_field:
|
||||
return datetime(2025, 1, 24)
|
||||
|
||||
# Default fallback
|
||||
return datetime.now() + timedelta(days=1)
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error parsing date: {e}")
|
||||
return datetime.now() + timedelta(days=1)
|
||||
|
||||
async def fetch_weather_data(self, latitude: float, longitude: float,
|
||||
target_date: datetime, days_range: int = 3) -> Dict:
|
||||
"""Fetch structured weather data for specific coordinates and date range."""
|
||||
|
||||
# Calculate date range around target date
|
||||
start_date = target_date - timedelta(days=days_range//2)
|
||||
end_date = target_date + timedelta(days=days_range//2)
|
||||
|
||||
params = {
|
||||
"latitude": latitude,
|
||||
"longitude": longitude,
|
||||
"current": [
|
||||
"temperature_2m",
|
||||
"relative_humidity_2m",
|
||||
"wind_speed_10m"
|
||||
# "wind_direction_10m",
|
||||
# "weather_code",
|
||||
# "pressure_msl",
|
||||
# "cloud_cover"
|
||||
],
|
||||
"hourly": [
|
||||
"temperature_2m",
|
||||
"relative_humidity_2m"
|
||||
# "wind_speed_10m",
|
||||
# "wind_direction_10m",
|
||||
# "precipitation",
|
||||
# "weather_code",
|
||||
# "visibility",
|
||||
# "cloud_cover",
|
||||
# "pressure_msl"
|
||||
],
|
||||
# "daily": [
|
||||
# "temperature_2m_max",
|
||||
# "temperature_2m_min"]
|
||||
# # "sunrise",
|
||||
# "sunset",
|
||||
# "precipitation_sum",
|
||||
# "wind_speed_10m_max",
|
||||
# "wind_direction_10m_dominant"
|
||||
# ],
|
||||
# "start_date": start_date.strftime("%Y-%m-%d"),
|
||||
# "end_date": end_date.strftime("%Y-%m-%d"),
|
||||
# "timezone": "auto"
|
||||
}
|
||||
|
||||
# Convert lists to comma-separated strings for API
|
||||
for key, value in params.items():
|
||||
if isinstance(value, list):
|
||||
params[key] = ",".join(value)
|
||||
|
||||
async with aiohttp.ClientSession() as session:
|
||||
async with session.get(self.base_url, params=params) as response:
|
||||
if response.status == 200:
|
||||
return await response.json()
|
||||
else:
|
||||
raise Exception(f"Weather API request failed: {response.status}")
|
||||
|
||||
def structure_weather_data(self, raw_weather: Dict, target_date: datetime) -> Dict:
|
||||
"""Structure raw weather data into organized format."""
|
||||
try:
|
||||
structured_data = {
|
||||
"location": {
|
||||
"latitude": raw_weather.get("latitude"),
|
||||
"longitude": raw_weather.get("longitude"),
|
||||
"elevation": raw_weather.get("elevation"),
|
||||
"timezone": raw_weather.get("timezone")
|
||||
},
|
||||
"current_conditions": raw_weather.get("current", {}),
|
||||
"hourly_forecast": {
|
||||
"times": raw_weather.get("hourly", {}).get("time", []),
|
||||
"temperature": raw_weather.get("hourly", {}).get("temperature_2m", []),
|
||||
"humidity": raw_weather.get("hourly", {}).get("relative_humidity_2m", []),
|
||||
"wind_speed": raw_weather.get("hourly", {}).get("wind_speed_10m", []),
|
||||
"wind_direction": raw_weather.get("hourly", {}).get("wind_direction_10m", []),
|
||||
"precipitation": raw_weather.get("hourly", {}).get("precipitation", []),
|
||||
"weather_code": raw_weather.get("hourly", {}).get("weather_code", []),
|
||||
"visibility": raw_weather.get("hourly", {}).get("visibility", []),
|
||||
"cloud_cover": raw_weather.get("hourly", {}).get("cloud_cover", []),
|
||||
"pressure": raw_weather.get("hourly", {}).get("pressure_msl", [])
|
||||
},
|
||||
"daily_forecast": {
|
||||
"dates": raw_weather.get("daily", {}).get("time", []),
|
||||
"temperature_max": raw_weather.get("daily", {}).get("temperature_2m_max", []),
|
||||
"temperature_min": raw_weather.get("daily", {}).get("temperature_2m_min", []),
|
||||
"sunrise": raw_weather.get("daily", {}).get("sunrise", []),
|
||||
"sunset": raw_weather.get("daily", {}).get("sunset", []),
|
||||
"precipitation_sum": raw_weather.get("daily", {}).get("precipitation_sum", []),
|
||||
"wind_speed_max": raw_weather.get("daily", {}).get("wind_speed_10m_max", []),
|
||||
"wind_direction_dominant": raw_weather.get("daily", {}).get("wind_direction_10m_dominant", [])
|
||||
},
|
||||
"target_date": target_date.strftime("%Y-%m-%d"),
|
||||
"data_retrieved_at": datetime.now().isoformat()
|
||||
}
|
||||
|
||||
return structured_data
|
||||
|
||||
except Exception as e:
|
||||
return {"error": f"Error structuring weather data: {e}"}
|
||||
|
||||
async def extract_booking_weather_data(self, booking_data: Dict) -> Dict:
|
||||
"""Main extraction function - returns structured data for external use."""
|
||||
try:
|
||||
# Extract all booking information
|
||||
booking_info = self.extract_booking_info(booking_data)
|
||||
|
||||
# Extract coordinates and target date
|
||||
latitude, longitude = self.extract_coordinates(booking_data)
|
||||
print(f"Latitude: {latitude}, Longitude: {longitude}")
|
||||
target_date = self.parse_target_date(booking_data)
|
||||
|
||||
# Fetch raw weather data
|
||||
raw_weather = await self.fetch_weather_data(latitude, longitude, target_date)
|
||||
|
||||
# Structure the weather data
|
||||
structured_weather = self.structure_weather_data(raw_weather, target_date)
|
||||
|
||||
# Return combined structured data
|
||||
return {
|
||||
# "booking_data": booking_info,
|
||||
# "coordinates": {
|
||||
# "latitude": latitude,
|
||||
# "longitude": longitude
|
||||
# },
|
||||
# "target_date": target_date.strftime("%Y-%m-%d"),
|
||||
"weather_data": structured_weather,
|
||||
"extraction_metadata": {
|
||||
"processed_at": datetime.now().isoformat(),
|
||||
"api_endpoint": self.base_url,
|
||||
"data_source": "open-meteo"
|
||||
}
|
||||
}
|
||||
|
||||
except Exception as e:
|
||||
return {
|
||||
"error": f"Data extraction failed: {e}",
|
||||
"job_id": booking_data.get("job_id", "unknown"),
|
||||
"processed_at": datetime.now().isoformat()
|
||||
}
|
||||
|
||||
# Example usage
|
||||
async def main():
|
||||
"""Demonstrate the data extraction system."""
|
||||
|
||||
# Sample booking data
|
||||
booking_form_input = {
|
||||
"job_id": "1043",
|
||||
"job_overview": {
|
||||
"job_number": "Job #1043",
|
||||
"site_name": "Hightower Solar Farm",
|
||||
"assigned_date": "January 23, 2025",
|
||||
"status": "Scheduled"
|
||||
},
|
||||
"site_information": {
|
||||
"site_name": "Hightower Solar Farm",
|
||||
"region": "North England",
|
||||
"full_address": "Grange Lane, Manchester M34 7TF",
|
||||
"gps_coordinates": {
|
||||
"latitude": "13.41° N",
|
||||
"longitude": "52.52 W"
|
||||
}
|
||||
},
|
||||
"timing": {
|
||||
"start_time": "09:00 AM",
|
||||
"end_time": "10:30 AM",
|
||||
"survey_duration": "30–45 mins",
|
||||
"buffer_time": "45 mins"
|
||||
},
|
||||
"form": {
|
||||
"asset_type": "Solar Farm",
|
||||
"system_size": "5.2 MW capacity, approximately 16,000 panels across 12 hectares",
|
||||
"survey_purpose": "Insurance assessment",
|
||||
"assigned_engineer": "David Wilson - 0161-555-0876",
|
||||
"contact_person": "Sarah Thompson",
|
||||
"contact_phone": "0161-555-0234"
|
||||
}
|
||||
}
|
||||
|
||||
# Initialize extractor
|
||||
extractor = DroneWeatherDataExtractor()
|
||||
|
||||
# Extract structured data
|
||||
result = await extractor.extract_booking_weather_data(booking_form_input)
|
||||
|
||||
# Display structured output
|
||||
print("=== STRUCTURED DATA EXTRACTION ===")
|
||||
print(json.dumps(result, indent=2))
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -1,48 +0,0 @@
|
||||
booking_form_input = {
|
||||
"job_id": "1043",
|
||||
"job_overview": {
|
||||
"job_number": "Job #1043",
|
||||
"site_name": "Hightower Solar Farm",
|
||||
"assigned_date": "January 23, 2025",
|
||||
"status": "Scheduled"
|
||||
},
|
||||
"site_information": {
|
||||
"site_name": "Hightower Solar Farm",
|
||||
"region": "North England",
|
||||
"full_address": "Grange Lane, Manchester M34 7TF",
|
||||
"gps_coordinates": {
|
||||
"latitude": "53.4408° N",
|
||||
"longitude": "2.2426° W"
|
||||
}
|
||||
},
|
||||
"timing": {
|
||||
"start_time": "09:00 AM",
|
||||
"end_time": "10:30 AM",
|
||||
"survey_duration": "30–45 mins",
|
||||
"buffer_time": "45 mins"
|
||||
},
|
||||
"form": {
|
||||
"asset_type": "Solar Farm",
|
||||
"asset_details": "null",
|
||||
"site_name": "Hightower Solar Farm",
|
||||
"site_location": "Grange Lane, Manchester M34 7TF, GPS: 53.4408° N, 2.2426° W",
|
||||
"system_size": "5.2 MW capacity, approximately 16,000 panels across 12 hectares",
|
||||
"access_type": "Restricted Access",
|
||||
"access_details": "Main gate access code: 7841. Contact facility manager Sarah Thompson at 0161-555-0234 for entry. Follow yellow markers to operations building. Site requires visitor registration.",
|
||||
"vehicle_access": "Yes - Vehicle Accessible",
|
||||
"contact_person": "Sarah Thompson",
|
||||
"contact_phone": "0161-555-0234",
|
||||
"contact_role": "Facility Manager",
|
||||
"contact_availability": "Available on-site",
|
||||
"special_safety_requirements": "Hard hats, high-vis vests, and steel-toe boots mandatory. Safety induction required (20 minutes). Keep 100m distance from substation area. Weather restriction: no flights in winds >15mph.",
|
||||
"survey_purpose": "Insurance assessment",
|
||||
"survey_purpose_details": "Annual insurance inspection required for policy renewal. Focus on structural integrity, panel condition, and inverter housing. Insurance company requires thermal imaging and detailed photographic documentation.",
|
||||
"additional_requirements": "Morning survey preferred for optimal lighting conditions. Need detailed documentation of any panel damage or hotspots. Provide georeferenced images for insurance mapping. Weather backup date: January 24th same time.",
|
||||
"timing_preference": "Within 1 week",
|
||||
"preferred_dates": "January 23rd morning preferred, backup January 24th",
|
||||
"selected_slot": "January 23rd, 2025 - 09:00 AM - 10:30 AM",
|
||||
"assigned_engineer": "David Wilson - 0161-555-0876",
|
||||
"booking_timestamp": "2025-01-15T10:45:22Z",
|
||||
"booking_id": "DSV-2025-0115-007"
|
||||
}
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
import os
|
||||
import json
|
||||
import asyncio
|
||||
from langchain_openai import ChatOpenAI
|
||||
from langchain_core.messages import HumanMessage, SystemMessage
|
||||
from src.prompts.templates.flght_prompt import flight_prompt # You must save the prompt function here
|
||||
from src.config.llm_config import LlmConfig
|
||||
from config import Config
|
||||
from src.components.data_extraction.weather_data import DroneWeatherDataExtractor
|
||||
|
||||
class DroneAssessmentAgent:
|
||||
def __init__(self):
|
||||
self.llm = ChatOpenAI(
|
||||
api_key=Config.OPENAI_API_KEY,
|
||||
model=LlmConfig.openai.models.gpt_4o,
|
||||
temperature=0.3
|
||||
)
|
||||
|
||||
self.weather_extractor = DroneWeatherDataExtractor()
|
||||
|
||||
async def run(self, booking_form: str) -> dict:
|
||||
"""
|
||||
Run the drone environmental & safety assessment agent.
|
||||
|
||||
Args:
|
||||
booking_form_text (str): Raw text of the booking form
|
||||
|
||||
Returns:
|
||||
dict: JSON output of the AI assessment
|
||||
"""
|
||||
weather_data = await self.weather_extractor.extract_booking_weather_data(booking_form)
|
||||
booking_form_text = json.dumps(booking_form)
|
||||
|
||||
prompt = flight_prompt(booking_form_text, weather_data) # Contains the instructions + JSON template
|
||||
messages = [
|
||||
SystemMessage(content=prompt),
|
||||
HumanMessage(content=booking_form_text)
|
||||
]
|
||||
|
||||
try:
|
||||
response = self.llm.invoke(messages)
|
||||
|
||||
if hasattr(response, "content"):
|
||||
response_text = response.content.strip()
|
||||
print("Raw LLM Response:\n", response_text[:300], "...\n")
|
||||
|
||||
# Attempt to parse JSON from response
|
||||
start_idx = response_text.find('{')
|
||||
end_idx = response_text.rfind('}') + 1
|
||||
|
||||
if start_idx == -1 or end_idx == -1:
|
||||
raise ValueError("No JSON object found in output")
|
||||
|
||||
json_str = response_text[start_idx:end_idx]
|
||||
return json.loads(json_str)
|
||||
|
||||
else:
|
||||
raise ValueError("LLM returned no usable content")
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error in DroneAssessmentAgent: {str(e)}")
|
||||
return {
|
||||
"error": str(e),
|
||||
"raw_response": response.content if 'response' in locals() else None
|
||||
}
|
||||
|
||||
async def main():
|
||||
"""Async main function to properly handle the async agent."""
|
||||
from test2 import booking_form_input
|
||||
|
||||
agent = DroneAssessmentAgent()
|
||||
result = await agent.run(booking_form_input)
|
||||
|
||||
print("\nStructured Assessment Output:\n")
|
||||
print(json.dumps(result, indent=2))
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
Reference in New Issue
Block a user