Public EV charging is rapidly expanding across outdoor parking environments, requiring operators to deploy more chargers across more locations without compromising performance or user experience. As infrastructure scales, the pressure is on to deliver systems that are faster to deploy, easier to manage, and reliable in real-world conditions.
To meet this demand, an intelligent EV charging bollard solution has been developed using edge AI, purpose-built for practical deployment in active parking environments. By processing key functions locally on the device, the system delivers a fast and consistent user experience, reducing the time from payment to charging start to under five seconds.
This approach has already proven its value in the field, enabling deployments of more than 1,000 bollards within a year while reducing maintenance and management costs by up to 30% compared to conventional outdoor charging systems.
Challenges
Outdoor EV charging installations present a unique set of challenges that go beyond standard infrastructure projects. Installations often take place in tight, high-traffic environments where bulky systems and complex wiring are not practical. Deployment windows are limited, requiring solutions that can be installed quickly and consistently, often during off-peak hours.
At the same time, user experience remains a critical factor. Drivers are far less likely to engage with charging stations if the process feels fragmented or slow. Cloud-reliant systems can also introduce delays in areas with unstable connectivity, leading to queues and reduced charger availability.
Key challenges include:
- Limited space and difficult installation conditions
- Short deployment timeframes
- Inconsistent power and network environments
- Fragmented or slow user authentication processes
Key Requirements
Addressing these challenges requires more than just hardware. EV charging systems need to be designed for both deployment efficiency and real-time operation.
This means combining a compact edge AI platform with the ability to process data locally, while also simplifying integration and maintaining reliable connectivity.
In practical terms, this includes:
- A compact system suitable for bollard integration
- Real-time AI processing without cloud dependency
- Simplified cabling and peripheral connectivity
- Reliable operation in outdoor environments
EV Charging Bollard Solution with Axiomtek AIE100-ONA
At the core of this solution is the Axiomtek AIE100-ONA, available through Tekdis. Designed for edge AI deployments, it combines high-performance computing with a compact, deployment-ready design.
Built on the NVIDIA Jetson Orin Nano platform, the system delivers up to 67 TOPS of AI performance, enabling real-time decision making directly at the point of service. This removes the delays associated with cloud processing and ensures consistent response times for users.
The platform also simplifies integration. Dual Ethernet with dedicated PoE support allows connected devices such as cameras to be powered and connected through a single cable, reducing internal complexity. Wireless auto-recovery helps maintain uptime, while the fanless, rugged design supports continuous operation across a wide temperature range.
Empowering EV Charging with Edge AI
By shifting intelligence to the edge, the charging bollard becomes an active system rather than just a connection point. It can process inputs locally and respond in real time, improving both efficiency and user experience.
This enables capabilities such as:
- License plate recognition for vehicle identification
- Parking duration tracking for billing and enforcement
- Real-time occupancy detection for bay availability
Because data is processed locally, only relevant outputs such as events and system status are transmitted upstream. This reduces bandwidth usage and ensures consistent performance, even in environments with unreliable connectivity.
AIE100-ONA Key Features
The AIE100-ONA is designed to balance performance with deployment practicality. It offers high AI compute capability, flexible I/O for integration, and features that support reliable, long-term operation in demanding environments.
Key highlights include:
- Up to 67 TOPS AI performance with NVIDIA Jetson Orin Nano
- Integrated PoE support for simplified device connectivity
- Smart wireless auto-recovery for improved uptime
- Wide operating temperature range from -25°C to +70°C
Why Tekdis
As an official Axiomtek distributor, Tekdis supports customers from evaluation through to deployment, providing access to proven edge AI platforms backed by local expertise.
With experience across industrial computing and AI applications, Tekdis helps simplify integration, reduce deployment time, and support scalable rollouts in real-world environments.
View Related Products
- Arm Cortex-A78AE v8.2 CPU
- NVIDIA Ampere architecture GPU
- 12V DC, -20°C ~ 50°C
