🤖 AI Planning Techniques for Robotics

Course: Planning Techniques for Robotics — 2025/2026 Author: Youssef Emad University: Benha University — Computer Science, AI Track

A collection of 3 optimization algorithms implemented in Python, applied to real-world planning and resource allocation problems.

---

📁 Repository Structure

Planning-Techniques/
│
├── 🧬 Genetic_Algorithm/
│   ├── Genetic_Algorithm_TSP.ipynb
│   └── README.md
│
├── 🐜 Ant_Colony_Optimization/
│   ├── ACO_QAP.ipynb
│   └── README.md
│
├── 🐝 Artificial_Bee_Colony/
│   ├── ABC_Bandwidth.ipynb
│   └── README.md
│
└── README.md  ← you are here

---

🔬 Algorithms Overview

#	Algorithm	Problem	Objective
1	🧬 Genetic Algorithm (GA)	Travelling Salesman Problem (TSP)	Minimize total travel distance
2	🐜 Ant Colony Optimization (ACO)	Quadratic Assignment Problem (QAP)	Minimize total assignment cost
3	🐝 Artificial Bee Colony (ABC)	Network Bandwidth Allocation	Maximize weighted user satisfaction

---

🧬 1. Genetic Algorithm — TSP

Problem: Find the shortest route visiting 4 Egyptian cities exactly once and returning to start.

Cities: Benha → Shebin Al-Kom → Al-Zagazig → Al-Mansoura → Benha

Parameter	Value
Population Size	50
Crossover Probability	0.82
Mutation Probability	0.30
Generations	30,000

Key Steps: Initialization → Fitness → Tournament Selection → Single-Point Crossover → Mutation → Repair

Result:

Benha → Al-Zagazig → Al-Mansoura → Shebin Al-Kom → Benha
Total Distance: 221 KM

📂 Genetic Algorithm

---

🐜 2. Ant Colony Optimization — QAP

Problem: Assign 5 facilities to 5 locations to minimize total interaction cost ∑ Distance × Flow.

Parameter	Value
Ants	30
Iterations	2000
Alpha (α)	1.2
Beta (β)	0.8
Rho (ρ)	0.26
Q	3

Key Steps: Ant Initialization → Probabilistic Transition → Fitness Evaluation → Pheromone Update

Result:

Assignment : [1, 4, 3, 2, 0]
Total Cost : 442

📂 Ant Colony Optimization

---

🐝 3. Artificial Bee Colony — Bandwidth Allocation

Problem: Allocate 750 Gbps across 6 clients to maximize weighted log-utility satisfaction Σ priority_i × ln(BW_i).

Parameter	Value
Bees	30
Iterations	2000
Limit	25
Penalty	10,000
Total BW Pool	750 Gbps

Key Steps: Employed Phase → Onlooker Phase → Elitism → Scout Phase

Result:

Total BW Used : 749.99 Gbps
Objective f(x): 81.6458
Constraint satisfied: True

📂 Artificial Bee Colony

📊 Comparison

Algorithm	Type	Problem Type	Key Mechanism
GA	Evolutionary	Combinatorial	Crossover + Mutation
ACO	Swarm Intelligence	Combinatorial	Pheromone Trails
ABC	Swarm Intelligence	Continuous	Employed / Onlooker / Scout

---

🛠️ Requirements

pip install numpy

All algorithms run with Python 3.x — only NumPy required (GA uses built-in random only).

---

📚 Topics Covered

• Genetic Algorithm (GA) · Ant Colony Optimization (ACO) · Artificial Bee Colony (ABC)
• Travelling Salesman Problem (TSP) · Quadratic Assignment Problem (QAP)
• Network Resource Allocation · Bio-inspired Optimization
• Swarm Intelligence · Evolutionary Computation