🚦 TLS Optimization using Genetic Algorithm and SUMO
The optimization minimizes the average vehicle waiting time and queue length, while maximizing throughput.
What is a Genetic Algorithm (GA)?
A Genetic Algorithm (GA) is an evolutionary optimization technique inspired by the process of natural selection in biology.
It is useful when the search space is large and complex — where traditional methods (like brute-force or gradient-based optimization) are not feasible.
Genetic Algorithm Steps
| Step | Process | Description |
|---|---|---|
| 1. Initialization | Create a random population | Each individual (chromosome) represents a possible traffic signal plan — e.g., [30, 3, 45, 4] for green/yellow durations. |
| 2. Fitness Evaluation | Evaluate each solution | Simulate each plan in SUMO and calculate its fitness (based on wait time, queue length, throughput). |
| 3. Selection | Choose the best solutions | Better-performing solutions have higher chances of being selected for reproduction. |
| 4. Crossover (Recombination) | Combine two solutions | Mix timing values from two parents to create new offspring with potentially better performance. |
| 5. Mutation | Randomly change some values | Introduce randomness (e.g., slightly changing a green duration) to maintain diversity and explore new possibilities. |
| 6. Replacement | Form a new population | Keep the best individuals (elitism) and replace the worst ones with new ones. |
| 7. Termination | Stop after certain generations | End when improvement slows or a maximum generation count is reached. |
Why Use a Genetic Algorithm for Traffic Optimization?
| Challenge | Why GA Helps |
|---|---|
| Dynamic and nonlinear environment | Traffic flow depends on multiple interacting factors (routes, timings, vehicle arrivals). GA can search through complex spaces without needing explicit formulas. |
| No single best timing plan | GA can adaptively find near-optimal signal timings even as traffic conditions change. |
| Multi-objective optimization | GA can balance competing objectives (e.g., reducing waiting time, minimizing queue length, maximizing throughput). |
In short: GA mimics natural evolution to continuously improve traffic light timings — leading to smoother traffic flow and reduced congestion.