expert system - automated traffic light control based on road congestion
TRANSCRIPT
AUTOMATIC TRAFFIC LIGHT
CONTROL BASED ON ROAD
CONGESTION
EXPERT SYSTEM
-Kartik Shenoy-
Motivation
Problems caused by traffic congestion:
• Missed opportunities, loss of time for commuters
• Lost worker productivity, trade opportunities, delivery delays, increased costs for employers
• Trouble to Traffic Police in coordinating and directing the traffic
Solutions possible:
• Improve road infrastructure
• Create new transport facilities
• Use technology to manage this congestion
Problem Definition
• Use video feed and loop detectors for managing traffic across multiple
intersections by controlling the traffic signals at these intersections.
• Aim:
• Minimize traffic congestion
• Maximize traffic flow
• Prevent traffic jams
• Reduce load on traffic police for handling traffic
Modules
Image Courtesy: Google Images
User Interface
• Humans drive cars (and follow traffic rules)
• Traffic signals are the main user interface for this expert system
• These indicate the user what to do next
• Knowledge Engineers may use server computer for changing fuzzy rules
Knowledge Base
• KB consists of
• rule based knowledge for deciding which signals to change and for what time to keep
it that way depending on inputs
• case specific knowledge as input to system
• Rules are stored as if <antecedent clauses> then <consequent clauses> rules
• Basic traffic rules are also stored
Example Rules used by [1]
• Rule: 1 if 3.0 < Interarrival_time then Singal_Type = ‘‘1’’
• Rule: 2 if 1.7 < Interarrival_time <= 3.0 then Singal_Type = ‘‘2’’
• Rule: 3 if 0 5 < Ineterarrival_time <= 1.7 then Singal_Type = ‘‘3’’
• Rule: 4 if Interarrival_time = ‘‘Exception’’ then Singal_Type = ‘‘4’’
• Rule: 5 if Singal_Type = ‘‘1’’ then Red_light_duration =65 and Green_light_duration = 95
• Rule: 6 if Singal_Type = ‘‘2’’ then Red_light_duration= 65 and Green_light_duration = 110
• Rule: 7 if Singal_Type = ‘‘3’’ then Red_light_duration =65 and Green_light_duration = 125
• Rule: 8 if Singal_Type = ‘‘4’’ then Red_light_duration =‘‘Manual’’ and Green_light_duration =
‘Manual’’
Case Specific Knowledge Acquisition
• Loop Detector or Video Detector or RFID[1] for finding NVWQ (No of Vehicles Waiting for Queue) [2]
• Video Feed for detecting accidents
• From this data at various intersections calculate maximum flow, inter arrival time, inter departure time, average car speed
• Here the system gathers the information automatically and humans don’t need to voluntarily provide data
Image Courtesy: Google Images
Image Courtesy: [1]
Inference Engine
• Case Specific KB (CSKB) acquisition –Receive data from loop detectors, video feeds and calculate inter arrival, departure times and NVWQ
• Fuzzy Controller[2] uses CSKB and temporal information (past flow) across various intersections to decide signal times and sequences across intersections
Image Courtesy: [2]
Image Courtesy: [4]
Image Courtesy: [6]
Image Courtesy: [1]
Simulation Model as used by [4]
• The agent receives at (given) time intervals the information on the current state of traffic (data collection).
• The agent receives information on other adjoining signalised intersections from other ITSA's (data collection).
• The agent has an accurate model of the controlled intersection and knows the traffic rules (analysis).
• The agent knows the recent trends (analysis/interpretation of data).
• The agent should be able to calculate the next cycle mathematically correct (analysis/decision).
• The agent should be able to actuate the next cycle and operate the signals accordingly (control).
• The agent should be able to detect and handle current traffic problems by itself (analysis/decision and control/action) and should inform other agents of the nature, severity and possible cause of the problem, if necessary (data distribution).
• The agent passes information on to other adjoining agents (data distribution).
Conclusion
• The accuracy of NVWQ estimation using the fuzzy neural networks
approaches is more than 90% [2]
Currently Used At
• Isolated Intersections Automatic Traffic Signal Control
• MOVA (UK)
• LHOVRA (Sweden)
LHOVRA[4]
• L: Freight Traffic - Detector 300m away
• H: Priority For Main Road - Detector 200m away
• O: Accident Reduction by Dilemma - Detector 140m away
• V: Variable Yellow Light - Yellow light retained if traffic continues to flow
• R: Red-light negative protection by prolonged evacuation time
• A: All Red function
References
[1] W. Wen, “A dynamic and automatic traffic light control expert system for
solving the road congestion problem”
[2] L. Conglin, W. Wu, IEEE Member, Tan Yuejin, “Traffic Variable Estimation and Traffic Signal Control Based on Soft Computation”, 2004
[3] K. W. Lim, G. C. Kim, “Knowledge-Based Expert System in Traffic Signal Control Systems”
[4] D. A. Roozemond, “Using Intelligent Agents For Urban Traffic Control Systems”
[5] https://nl.wikipedia.org/wiki/LHOVRA, Sweden
[6] A. Zaied, W. Othman, “Development of a fuzzy logic traffic system for isolated signalized intersections in the State of Kuwait”
Thank You