AI Search Algorithms for Smart Mobility#

Warning

This Jupyter book is no longer actively maintained. For updated and more extended version of these materials, please check Optimization Algorithms.

Smart mobility is the promotion of sustainable mobility that guarantees seamless access to different modes of mobility, and enables people or cargo to get from one place to another in a way that is safe, clean, and most efficient (fast, convenient, comfort, productive and affordable). This disruptive technology is people-centric, software-defined, connected, and electric [1]. There are several ill-structured optimization problems in smart mobility systems and services that cannot be solved using traditional search algorithms. These problems include, but are not limited to:

  • multi-criteria optimal routing

  • emergency dispatch and routing for first response emergency vehicles

  • self-driving vehicle motion planning

  • ridesharing, ride-hailing or ridesourcing

  • dynamic pricing

  • dynamic on-demand mobility services

  • multi-modal transportation planning

  • last-mile delivery systems using droids/cargo-bikes

  • deadheading

  • platooning or flocking and

  • fleet management

AI search algorithms have the power in dealing with complex discrete and continuous optimization problems in the smart mobility domain. Most of these techniques are nature-inspired, stochastic optimization methods that iteratively use random elements to transfer one candidate solution into a new, hopefully, better solution with regards to a given measure of quality.



Topics to be covered include:

  • geospatial data science

  • graph search algorithms (blind and informed search algorithms)

  • heuristics and metaheuristics

  • trajectory-based methods (tabu search and simulated annealing)

  • evolutionary computing methods (genetic algorithms and differential evolution algorithm)

  • swarm intelligence algorithms (particle swarm optimization, ant colony optimization, artificial bee colony algorithm, firefly algorithm, and stochastic diffusion search)

  • parallel and hybrid metaheuristics

  • learn to search (geometric deep learning, graph neural networks, attention mechanisms and deep reinforcement learning)

The book provides examples and in-depth case studies to show the ability of the covered search algorithms in solving people mobility, logistics, and infrastructure optimization problems. Python implementations in the form of Jupyter notebooks are provided and available through the book’s website on GitHub. AI Search Algorithms for Smart Mobility is written mainly as a project-oriented practical book intended for academic institutions, continuing education, training centers, and working professionals.