Extending EV Range with AV Programming

Intelligent programming of autonomous electric vehicles offers potentially big energy savings, according to a study by IAV.

Optimizing an EV’s automated-driving potential can produce up to a 50% gain in real-world urban-driving range and up to 13% in highway use. (Tesla)

Increasing the range of electric vehicles (EV) in a practical cost-effective way has become a primary challenge for the industry. However, current efforts to increase EVs’ market acceptance, via advertising, typically focus on the exhilaration that comes from mashing the accelerator and enjoying the launch torque compared with a gasoline engine. An alternative approach, with customer appeal of its own, would be to increase the vehicle’s autonomous driving capability.

This was proposed by Jonas Gutsche, an advanced-development engineer with IAV, at the recent SAE Thermal Management Systems Symposium. He said the vehicle could produce real-world gains in range, perhaps 50% in urban driving and 13% in highway use. Most battery electric vehicles (BEV) provide a choice of driving styles (economy, normal and sportive), but the mode selected is still up to the driver.

HAV city cycle simulation results (up to 29.3% less energy used) were combined with regenerative braking to indicate a total of about 50% savings. (IAV)

Three aspects of an AV trip

The IAV study outlined by Gutsche looked at the three aspects of a trip for a “highly automated vehicle” (HAV)—essentially SAE autonomous levels 4 to 5. The first, pre-processing, uses the map data to determine the route. The second aspect, “Macro” factors, must be considered along with the pre-processing. They include integration of the details from the route itself—speed limits, required turns, intersections and time factors.

The pre-programming also must consider the driving dynamics and driving/rolling resistance (drag) of the vehicle, and the changes in the battery pack state of charge. As an example, this last item is affected by and itself affects everything from route selection to availability and usability of regenerative braking and any other recuperative system in the car.

Highway cycle simulation testing also showed worthwhile range improvement with autonomous control, while performing such “real world” maneuvers as overtaking slow-moving trucks. (IAV)

The third aspect is how to adjust for the micro factors – “while driving” inputs that often change instantly – not precisely predictable during route planning. They include any car-to-infrastructure transmissions (including from traffic control devices), sensor data and camera readings from on-board devices and weather changes.

Gutsche pointed out the most efficient deceleration is coasting – no active reduction in speed and no waste of energy. But unless a battery pack’s state of charge (SoC) is below 90%, there won’t be energy recovery from regeneration, and in-city driving presents fewer coasting opportunities. In his SAE presentation, Gutsche offered an energy-efficient urban acceleration-and-braking approach, noting that if conditions don’t permit the calculated drive, the vehicle should be operated for maximum efficiency (slow, light acceleration) until opportunities for the anticipated strategies emerge.

Regenerative braking (recuperation) is particularly significant in “real world” simulations on city cycle with many turns and bends in the route. (IAV)

Value in BEV “driver training” algorithms

The IAV simulation results that showed maximum improvement with HAV control (50%) were compared to “sportive” urban driving. A “moderate” human driving style could approach the HAV results according to the IAV study, indicating value in driver-training algorithms available in some BEVs. However, the computer still delivered a significant improvement over the moderate driver via regenerative braking, vs. none for the sportive driver.

Although the range increase was not as dramatic for the HAV in a two-lane highway driving comparison, it still was worthwhile, ranging from 6.5% to 13.3%. Two simulations were run, one at 120 km/h (72 mph), a second at 150 km/h (90 mph). Four trucks traveling at 80 km/h (48 mph) were programmed into the simulation, with the HAV overtaking them. In both cases, there was worthwhile recuperation (regenerative braking) derived from the HAV, none from the sportive driver and only a minor amount of “regen” from the moderate driver.

Gutsche said IAV’s next steps are to get real test data by applying this simulation and the data it produced to an IAV-developed HAV, while also incorporating range-enhancing strategies for thermal management and auxiliary components.