Aetina launches palm-sized in-vehicle edge AI systems

Aetina has launched its DeviceEdge AIE-VN34/44 and AIE-VO24/34, described by the company as its first palm-sized in-vehicle edge AI systems, alongside a preview of a larger, ruggedised AIE-VX28/18-IP platform.

The AIE-VN34/44 and AIE-VO24/34 are powered by the NVIDIA Jetson Orin NX and NVIDIA Jetson Orin Nano modules running in Super Mode, delivering up to 100 TOPS of AI inference performance. Each system supports up to four GMSL2 camera inputs. GMSL2 (Gigabit Multimedia Serial Link 2) is a serializer-deserializer (SerDes) interface used to carry high-bandwidth video and control data over a single cable, reducing cabling weight and complexity in vehicle and robot installations. Aetina positions the two systems for multi-camera perception in vehicles, autonomous mobile robots (AMRs) and industrial equipment.

Aetina is also previewing the AIE-VX28/18-IP, an IP66-rated, fanless in-vehicle system built around the NVIDIA Jetson AGX Orin module. The platform supports eight GMSL2 camera inputs plus four M12 PoE camera inputs, 10GbE connectivity, and carries E-Mark (E24) certification. Aetina describes the design as intended for logistics fleets, intelligent transportation systems and other outdoor mobility deployments requiring sealed, fanless enclosures.

The same AIE-VX28/18-IP hardware appears in a separate demonstration Aetina plans to show at Smart City Expo World Congress (SCEWC) 2026 in Barcelona, from 3 to 5 November, where truck-mounted cameras feed the unit running NVIDIA DeepStream on NVIDIA CUDA to process eight concurrent camera streams for loading-dock safety detection. Aetina says that with NVIDIA JetPack 7.2 enabling Super Mode on the Jetson AGX Orin 32GB module, AI performance in that configuration rises from 200 to 241 TOPS, which the company says provides additional headroom for more complex vision pipelines without moving to a higher-specification module.