ENHANCING CAMPUS NAVIGATION: A VRP ANALYSIS FOR AUTONOMOUS VEHICLE ROUTING

Kavish Shah

Download Paper | Permalink

Conference: 2025 ASEE PSW Conference
Location: California Polytechnic University, California
Publication Date: April 10, 2025
Start Date: April 10, 2025
End Date: April 12, 2025
DOI: 10.18260/1-2--55172
Permanent URL: https://peer.asee.org/55172

Paper Authors

author page

Kavish Shah California State Polytechnic University, Pomona

Download Paper | Permalink

Abstract

There are many obstacles to overcome when integrating autonomous vehicles (AVs) into daily life, especially when it comes to real-time route optimization. Cal Poly Pomona plans to implement AVs for campus-wide internal transportation to improve mobility and offer faculties, staff, and students a cutting-edge, effective transit option. To increase travel efficiency while reducing time and distance, this study explores the Vehicle Routing Problem (VRP) as a potential solution for optimizing AV routes inside a university campus setting. The main objective of this study is to create a VRP-based algorithm that is specially designed to optimize AV routes on campuses. This study's particular goals are as follows: 1. Developing and Testing a Real-Time Route Optimization Algorithm: This algorithm will use real-time data to generate the best possible routing options that are flexible enough to adjust to the changing conditions of the route. 2. Assessing the Effectiveness of Various Algorithmic Techniques: We will investigate a range of techniques, such as genetic and brute-force algorithms. 3. External Data Sources: By using APIs like OpenRouteService, we hope to improve route computations by adding real-time traffic information and road conditions, allowing for more precise and flexible route planning.

The Traveling Salesman Problem (TSP), a subset of VRP, addresses optimizing routes for vehicles servicing multiple destinations. In this study, we apply the concepts of VRP to the field of AV navigation. The main challenge is choosing the best routes while taking into consideration dynamic elements like traffic patterns, road closures, and other real-time limitations. Innovative strategies that make use of modern technology and computational methods are required since existing solutions frequently find it difficult to adjust to the complexity of real-world routing, especially in settings with changing traffic patterns. The growing need for effective and sustainable urban mobility solutions highlights the significance of this study. A multi-step process is used in this study to guarantee thorough assessment and advancement of the suggested algorithm:

1. Algorithm Development: To create a baseline for testing and validation, a Python-based algorithm was first put into practice using the brute-force method. Despite being computationally demanding, this method made it possible to fully comprehend route development in a controlled environment. We now intend to apply heuristic techniques, like genetic algorithm, and compare the outcomes with those of brute-force technique. 2. Data Integration: Test scenarios were established using specific campus locations as a basis. To obtain the coordinates, access real-time traffic information and route computations, and improve the algorithm's adaptability to actual conditions, the OpenRouteService API was used to geocode the location names.

We defined key performance indicators (KPIs) such as total trip time, route efficiency (in kilometers), and computing time for route design to assess the efficacy of the suggested method. The feasibility of the suggested approach for real-time applications will be shown by benchmark testing versus traditional routing techniques. In addition to highlighting the algorithm's advantages, this assessment will point up possible areas for development. This study offers important information on how to use the VRP to optimize AVs navigation. The work advances the larger objectives of efficient urban mobility and sustainable transportation by implementing sophisticated routing algorithms on a university campus. To provide scalability for larger urban applications, ongoing work attempts to optimize the algorithms for a variety of traffic scenarios. Our research will help guide future advancements in autonomous navigation as AV technology advances, ultimately improving campus mobility and supporting the sustainable transportation ecosystem.

Citation
Format

Shah, K. (2025, April), ENHANCING CAMPUS NAVIGATION: A VRP ANALYSIS FOR AUTONOMOUS VEHICLE ROUTING Paper presented at 2025 ASEE PSW Conference, California Polytechnic University, California. 10.18260/1-2--55172

TY - CPAPER
AB - There are many obstacles to overcome when integrating autonomous vehicles (AVs) into daily life, especially when it comes to real-time route optimization. Cal Poly Pomona plans to implement AVs for campus-wide internal transportation to improve mobility and offer faculties, staff, and students a cutting-edge, effective transit option. To increase travel efficiency while reducing time and distance, this study explores the Vehicle Routing Problem (VRP) as a potential solution for optimizing AV routes inside a university campus setting.
The main objective of this study is to create a VRP-based algorithm that is specially designed to optimize AV routes on campuses. This study&#39;s particular goals are as follows:

1. Developing and Testing a Real-Time Route Optimization Algorithm: This algorithm will use real-time data to generate the best possible routing options that are flexible enough to adjust to the changing conditions of the route.
2. Assessing the Effectiveness of Various Algorithmic Techniques: We will investigate a range of techniques, such as genetic and brute-force algorithms.
3. External Data Sources: By using APIs like OpenRouteService, we hope to improve route computations by adding real-time traffic information and road conditions, allowing for more precise and flexible route planning.

The Traveling Salesman Problem (TSP), a subset of VRP, addresses optimizing routes for vehicles servicing multiple destinations. In this study, we apply the concepts of VRP to the field of AV navigation. The main challenge is choosing the best routes while taking into consideration dynamic elements like traffic patterns, road closures, and other real-time limitations. Innovative strategies that make use of modern technology and computational methods are required since existing solutions frequently find it difficult to adjust to the complexity of real-world routing, especially in settings with changing traffic patterns. The growing need for effective and sustainable urban mobility solutions highlights the significance of this study.
A multi-step process is used in this study to guarantee thorough assessment and advancement of the suggested algorithm:

1. Algorithm Development: To create a baseline for testing and validation, a Python-based algorithm was first put into practice using the brute-force method. Despite being computationally demanding, this method made it possible to fully comprehend route development in a controlled environment. We now intend to apply heuristic techniques, like genetic algorithm, and compare the outcomes with those of brute-force technique.
2. Data Integration: Test scenarios were established using specific campus locations as a basis. To obtain the coordinates, access real-time traffic information and route computations, and improve the algorithm&#39;s adaptability to actual conditions, the OpenRouteService API was used to geocode the location names.

We defined key performance indicators (KPIs) such as total trip time, route efficiency (in kilometers), and computing time for route design to assess the efficacy of the suggested method. The feasibility of the suggested approach for real-time applications will be shown by benchmark testing versus traditional routing techniques. In addition to highlighting the algorithm&#39;s advantages, this assessment will point up possible areas for development.
This study offers important information on how to use the VRP to optimize AVs navigation. The work advances the larger objectives of efficient urban mobility and sustainable transportation by implementing sophisticated routing algorithms on a university campus. To provide scalability for larger urban applications, ongoing work attempts to optimize the algorithms for a variety of traffic scenarios. Our research will help guide future advancements in autonomous navigation as AV technology advances, ultimately improving campus mobility and supporting the sustainable transportation ecosystem.
AU - Kavish Shah
CY - California Polytechnic University, California
DA - 2025/04/10
PB - ASEE Conferences
TI - ENHANCING CAMPUS NAVIGATION: A VRP ANALYSIS FOR AUTONOMOUS VEHICLE ROUTING
UR - https://peer.asee.org/55172
DO - 10.18260/1-2--55172
ER -