skip to main content
:::
:::

R&D Focus

Make Electric Scooters Smarter

The Self-Learning Dynamic Power Estimation Technology incorporates AI and enables an accurate display of remaining battery value and estimated mileage.

The Self-Learning Dynamic Power Estimation Technology incorporates AI and enables an accurate display of remaining battery value and estimated mileage.

Always worried that your electric scooter’s battery will run out of juice halfway to your destination? Or not believe in the battery level shown on the dashboard? These concerns will now be a thing of the past with the launch of the Self-Learning Dynamic Power Estimation Technology, an innovation co-developed by ITRI and KYMCO. Earlier this year, the two parties signed an agreement on electric scooter energy management systems that covers cooperation on 36 patented technologies to develop next generation AI-powered electric scooters.

The Self-Learning Dynamic Power Estimation Technology is developed due to the fact that the longer a battery for an electric scooter is in use, the greater the error in power forecasting. To alleviate riders’ anxiety about how much distance they can expect from the battery, ITRI’s technology incorporates AI and enables an accurate display of remaining battery value and estimated mileage. Riders can thus get rid of the habit of conserving power and avoid mileage loss.

The Self-Learning Dynamic Power Estimation Technology features three main characteristics:

1. Import of AI

The built-in power management AI system in electric scooters operates on the basis of what it has learned from different riding habits of scooter owners, providing the precise mileage and motor output accordingly, thereby greatly enhancing vehicle safety and reliability.

2. Display of accurate battery power and remaining mileage value

Unlike the imprecise battery percentage displayed in current electric scooters, ITRI’s technology shows actual battery value and provides estimations of remaining mileage expected from the battery, incorporating power consumption analysis. The technology can accurately estimate remaining mileage to within 5% regardless of the age of the battery.

3. Power failure alert

Battery discharge is limited in overly high or low temperature environments when a battery’s remaining charge is low. The system will proactively display the horsepower limit and will maintain a safe state for the vehicle to avoid triggering abnormal power failure protection. Even when a battery is running low or when start-up protection occurs, a power-off alarm message will be displayed before power is cut. This avoids sudden braking and other safety issues that are experienced in current electric scooters when batteries are depleted.

ITRI and KYMCO signed an agreement to collaborate on the development of electric scooter energy management systems earlier this year.

ITRI and KYMCO signed an agreement to collaborate on the development of electric scooter energy management systems earlier this year.

ITRI Executive Vice President Alex Y.M. Peng remarked that the technology developed by ITRI provides precise forecasts as to how far an electric scooter can travel on its battery, and the range of error in the prediction is less than 5% no matter if the scooter is operated in tropical or temperate regions. In addition, the battery can be monitored on a remote basis. When a rider is driving outside, the scooter company can monitor the health of the battery. This differentiates the product and gives it a competitive advantage, he said. The cooperation between ITRI and KYMCO marks the beginning of a new era, and it is hoped that the smart battery technology will help bring the Taiwan brand to the world, Dr. Peng added.

KYMCO President Ting-Yi Wang stated that the cooperation between his company and ITRI dates back over 20 years. Recently, the two have worked to strengthen power systems, control systems, motor cooling systems, and anti-theft systems. These enhancements are being used to bring their electric scooters to AI level.

Dr. Tzong-Ming Lee, ITRI Vice President and General Director of the Material and Chemical Research Laboratories, commented that the Self-Learning Dynamic Power Estimation Technology puts riders’ minds at ease so that they can drive greater distances. At present, test scooters equipped with this technology have been developed, and he is confident that the actual production of a scooter built with this technology in cooperation with KYMCO will offer consumers an entirely new experience.

The Self-Learning Dynamic Power Estimation Technology is applicable in all electric vehicles and energy storage systems. It has already been introduced in electric buses, high-voltage energy storage and other large-scale battery systems that require high safety and reliability.

TOP