From Propane to Pi: Affordable AI-Driven Solutions for Sustainable Bus Heating.
2026-01-0164
To be published on 04/07/2026
- Content
- The design of an affordably luxurious school bus means improving passenger comfort while greatly decreasing energy usage and emissions. The key consideration is to replace traditional propane heating with an electric heated-seat system and supported by green electricity/energy efficient smart power management. The system is designed from a 24V/48V battery system; to a Smart Energy Cube - a portable dock in power module requiring solar, battery and grid connection. Seat heating is only applied to seats that are occupied. The on-board image processing models (YOLOv5/YOLOv8, OpenCV, TensorFlow Lite) are run on affordable/portable microcomputers capable of detecting the number of occupied and available seats e.g. Raspberry Pi. There is one bottle neck when using the Raspberry Pi in the experiments to validate concept, when using low/medium power frame rates of ~ 1 fps for Raspberry Pi 3; to 7-8 fps with a Raspberry Pi 5. As a baseline, there is still potential for improvement and optimization opportunities. The energy analysis identifies that heating 72 seats only used 2.52 kW, while the propane heaters used about 10 kW, indicating a 75% demand reduction. The capital cost of heated seats is roughly about $1,900 per bus, while propane heating typically costs over $3,000 yearly in fuel. Over the 5 years examined, the potential saving of operating costs from the proposed seats was $10,000–12,000 per bus and reduced maintenance. The heated seats could be combined with solar and grid-tied systems increasing efficiency and utilizing bi-directional power. The combination of selective heated-seat technology with economical microcomputer-based control and scalable energy modules represents a realistic and affordable path for schools to modernize their transportation. This solution reduces heating energy use by over 70%, saves thousands of dollars per bus, while reducing emissions and offers a sustainable solution for large-scale adoption in school districts.
- Citation
- Chikkala, Daney Bhargav et al., "From Propane to Pi: Affordable AI-Driven Solutions for Sustainable Bus Heating.," SAE Technical Paper 2026-01-0164, 2026-, .