Robo-Uber Project: Building an Automated Taxi Dispatch System
Project Goal: Develop a simplified, intelligent system for an automated taxi service involving both taxi agents and a central dispatcher, focusing on optimization, planning, and bid-based dispatching.
Project Overview
In Robo-Uber, you’ll develop a multi-agent system where both taxis and a central dispatcher work collaboratively to transport users (‘fares’) efficiently. Taxis operate on a pre-set map, navigate around obstacles, and are responsible for bidding on fares while a dispatcher manages fare allocations and maximizes system returns.
System Components
- Service Area: A networked map of roads with random fare requests.
- Central Dispatcher: Handles fare requests, bidding, and fare assignments.
- Taxi Agents: Simulated taxis that handle route planning, bidding, and earning management.
Time in this simulation advances in one-second intervals, and traffic conditions, road capacity, and fare requests vary.
Project Tasks
Task 1: Path Planning (20%)
- Initial Evaluation:
- Run a basic system simulation and analyze taxi and dispatcher returns, operational uptime, and fare cancellations.
- Path Planning Enhancement:
- Modify the
_planPath
function to optimize route efficiency. Justify your approach and compare results with the baseline.
- Probabilistic Path Planning (Optional):
- Implement a path planner considering real-time traffic estimates and analyze its impact on fare cancellations.
Task 2: Dispatcher Optimization (10%)
- Enhance the
_allocateFare
function to maximize returns by optimizing fare scheduling, reducing wait times, and balancing bid requirements.
- Perform a comparative analysis of schedules and returns before and after changes.
Task 3: Taxi Bidding Strategy (20%)
- Bid Optimization:
- Update the
_bidOnFare
function to improve expected returns, comparing outcomes with and without traffic.
- System Analysis:
- Evaluate the commercial feasibility of this automated taxi service and areas needing further testing.
- Dispatcher Cost Function (Optional):
- Modify
_costFare
to maximize fare transport probability, minimize cancellations, and discuss its application in real-world deployments.
Assessment Criteria
Your project will be assessed on:
- Task Functionality: Meeting basic requirements for each component.
- Quality and Originality: Innovative approaches in planning, dispatch, and bidding.
- Coding Practice and Documentation: Clean, well-documented code.
- Evaluation and Analysis: Insightful comparisons and recommendations.
This individual project emphasizes foundational AI principles in transport logistics, providing practical experience in route optimization, fair bidding, and dispatch decision-making for real-time applications.