![flux_krea_00365_](https://cdn-uploads.huggingface.co/production/uploads/68401f649e3f451260c68974/Er62mlciwE7gxap16wOk_.png)

---
license: mit
tags:
- pathfinding
- simulation
- reinforcement-learning
- pyqt5
- autonomous-agents
- ai-education
- vacuum-cleaner
- search-algorithms
- bfs
- a-star
- manhattan-distance
- euclidean-distance
- chebyshev-distance
- turn-cost
- performance-metrics
- >-
  - turn-cost - performance-metrics - algorithm-comparison - visualization-tool
  - educational-software - python - artificial-intelligence -
  robotics-simulation - grid-world - obstacle-avoidance -
  multi-algorithm-framework - heuristic-evaluation - computational-efficiency -
  node-expansion - path-optimization - interactive-learning -
  real-time-simulation - gui-application - academic-project - research-tool -
  algorithm-visualization - performance-analysis - turn-penalty - cost-analysis
  - exploration-strategies - search-techniques - autonomous-navigation -
  intelligent-agents - simulation-environment
- >-
  - path-planning - heuristic-search - comparative-analysis -
  educational-resource - ai-simulation - robotics-education -
  algorithm-benchmarking - performance-metrics - visualization-framework -
  interactive-demonstration - learning-tool - academic-resource -
  simulation-software - ai-visualization - pathfinding-algorithms -
  search-methods - heuristic-functions - turn-cost-modeling -
  performance-evaluation - algorithm-testing - simulation-platform -
  educational-application - ai-demonstration - robotics-simulation -
  autonomous-systems - intelligent-systems - search-strategies -
  path-optimization - performance-comparison - heuristic-performance -
  algorithm-efficiency - simulation-tool - visualization-software -
  educational-software - ai-education - robotics-education -
  pathfinding-visualization - algorithm-visualization -
  performance-visualization - turn-cost-visualization -
  multi-algorithm-comparison - interactive-simulation - real-time-visualization
  - gui-simulation - pyqt5-application - python-simulation - grid-simulation -
  obstacle-navigation - dirty-cell-cleaning - autonomous-cleaning -
  vacuum-simulation - search-algorithm-comparison - heuristic-comparison
---
---
title: Vacuum Cleaner Search Simulation
emoji: 🏠
colorFrom: blue
colorTo: green
pinned: false
license: mit
---

# Vacuum Cleaner Search Simulation

An interactive simulation that demonstrates various search algorithms for vacuum cleaner pathfinding in a grid environment.

## 🎯 Overview

This application visualizes how different search algorithms navigate through a grid to find and clean dirty cells while avoiding obstacles. The simulation compares the performance of BFS and A* search with different heuristics.

## 🧠 Features

- **Multiple Search Algorithms**:
  - BFS (Breadth-First Search)
  - A* with Manhattan distance heuristic
  - A* with Euclidean distance heuristic  
  - A* with Chebyshev distance heuristic

- **Interactive Controls**:
  - Reset environment
  - Step-by-step simulation
  - Auto-run mode
  - Turn cost toggle (adds cost for direction changes)

- **Real-time Metrics**:
  - Steps taken and total cost
  - Nodes explored and expanded
  - Computation time
  - Algorithm performance comparison

## 🎮 How to Use

1. **Setup the Environment**:
   - The grid automatically generates with obstacles (blue) and dirty cells (red)
   - The vacuum starts at a random clean position

2. **Choose Algorithm**:
   - Select from the dropdown menu (BFS, A* Manhattan, A* Euclidean, A* Chebyshev)

3. **Configure Options**:
   - Toggle "Turn Cost" to enable/disable penalty for direction changes
   - Turn cost adds +0.5 for each 90° direction change

4. **Run Simulation**:
   - Click **Next** to advance one step
   - Click **Run** for continuous automatic execution
   - Click **Stop** to pause automatic execution
   - Click **Reset** to generate a new environment

## 🏗️ Technical Details

### Search Algorithms

- **BFS**: Explores all directions equally, guarantees shortest path
- **A* Manhattan**: Uses city-block distance heuristic (`|x1-x2| + |y1-y2|`)
- **A* Euclidean**: Uses straight-line distance heuristic
- **A* Chebyshev**: Uses chessboard distance heuristic (`max(|x1-x2|, |y1-y2|)`)

### Cell Types

- 🟡 **Yellow**: Clean cells
- 🔴 **Red**: Dirty cells (need cleaning)
- 🔵 **Blue**: Obstacles (block movement)
- 🟢 **Green**: Explored cells
- 🟠 **Orange**: Current path

### Performance Metrics

- **Nodes Explored**: Total unique positions visited
- **Nodes Expanded**: Total nodes processed by algorithm
- **Computation Time**: Time taken to find paths
- **Turn Cost**: Additional cost from direction changes

## 📊 Algorithm Comparison

The application tracks and compares:
- Average nodes expanded per run
- Average computation time
- Efficiency across different heuristics

Typically:
- **BFS** explores more nodes but finds optimal paths
- **A* Euclidean** is often most efficient for direct paths
- **A* Chebyshev** may explore more nodes in grid environments

## 🚀 Local Development