Hybrid Car Engineers Look to Flocking Birds to Hone Fuel Efficient Algorithms
【Summary】It involves creating vehicular algorithms that mimic the way a group of birds or swarms of bees process their surrounding environment. This is part of a new, emerging niche called vehicular science, which is currently dominated by Tesla, Mobileye and Google (no surprise here).
The natural design of animals and structures help engineers build more efficient products and systems that are deployed in real-world scenarios or environments. This can be seen in drones, satellites and commercial airplanes. In the sustainable car industry, engineers are applying the same framework to improve fuel efficiency in hybrid vehicles.
This nascent practice first surfaced in a published study titled Development and Evaluation of an Evolutionary Algorithm-Based Online Energy Management System for Plug-In Hybrid Electric Vehicles (IEEE Transactions on Intelligent Transportation Systems). It involves creating vehicular algorithms that mimic the way a group of birds or swarms of bees process their surrounding environment. This is part of a new, emerging niche called vehicular science, which is currently dominated by Tesla, Mobileye and Google (no surprise here).
"By mathematically modeling the energy saving processes that occur in nature, scientists have created algorithms that can be used to solve optimization problems in engineering," study author Xuewei Qi, a postdoctoral researcher at the Center for Environmental Research and Technology in UCR's Bourns College of Engineering said in a release.
The Learning Process
In order to understand how the algorithm works, one must first know about challenges surrounding fuel efficiency in hybrid cars, specifically plug-in variants. The units operate on stored energy via battery packs, as well as standard fuel (gasoline). While driving, the vehicle typically uses electric power, but also has the option to switch to fuel. For such applications, battery power is prioritized over fuel, with the latter serving as an extension of power provided from the power cells. Without a reliable system that "tells" the car when to make the switch, this protocol is notoriously inefficient. The majority of plug-in hybrid cars in the market today do not have features that prioritize fuel/power economy during operation.
This is the issue that UCR researchers are addressing with the AI-powered algorithm. It helps hybrid cars learn about their environment by leveraging the owner's driving history and using other driving data from an online network.
By implementing evolutionary algorithms in plug-in hybrid cars, researchers believe that the vehicles will be able to calculate and determine when it is appropriate to use electric power and when drivers should switch to standard fuel. Factors that may affect this decision (made autonomously or manually) includes traveling speed, road congestion and etc.
This technology is beneficial for individuals who previously purchased hybrid cars with the intention of reducing harmful emissions. When applied to a fleet of plug-in hybrid vehicles, fuel savings can reach monumental amounts. UCR researchers used the algorithm with other smart car technologies and were able to achieve energy savings as high as 30 percent. That's not bad at all, considering that in application, switching between electric power and fuel is often completed in a discreet manner.
"We combined this approach with connected vehicle technology to achieve energy savings of more than 30 percent. We achieved this by considering the charging opportunities during the trip—something that is not possible with existing EMS (energy management systems)," explained UCR scientists.
Michael Cheng is a legal editor and technical writer with publications for Blackberry ISHN Magazine Houzz and Payment Week. He specializes in technology business and digesting hard data. Outside of work Michael likes to train for marathons spend time with his daughter and explore new places.
Startup Valqari Reinvents Mailboxes to Cater to Drone Deliveries
Einride’s T-pod L4 Driverless Trucks Hit Public Roads in Sweden
Highways England Tests Autonomous Trucks for Road Construction
Drivent Reveals Self-driving Safety Products, No Longer in Stealth Mode
Amazon Reportedly Considering Acquisition of Driverless Startup TuSimple
Ford Plans to Deploy 100 Self-driving Vehicles in 2019
Amazon Unleashes In-car Delivery Service to Ford and Lincoln Vehicles
Washington State Regulators Move Forward to Allow Delivery Bots on Sidewalks
- EVs Expected to Account For Half of New-Car Sales by 2040
- Tesla Gigafactory 3 Construction in Shanghai is 'Ludicrously Fast'
- Hyundai’s Upcoming App Allows EV Owners to Adjust Their Car’s Performance
- Ford Will Produce its New Electric Models in Michigan & Chicago
- Tesla’s Long Range Model S Now Has a Range of 370 Miles
- May Mobility to Begin Testing Autonomous Shuttles in Rhode Island
- Reuters Poll Finds American Drivers Still Aren’t Interested in Autonomous Cars
- Baidu Tops the List for the Most Autonomous Miles Driven in China
- Singapore Develops Driverless Solutions for the Elderly
- General Motors Looking to Refine Super Cruise Before 2020 Expansion