Inbolt Introduces On-Arm AI Vision for Industrial Bin Picking

Inbolt, global industry experts in real-time 3D vision and AI-guided robotic perception, has announced the launch of a next-generation industrial bin picking solution designed for unstructured manufacturing environments. The system uses an on-arm 3D vision architecture combined with proprietary artificial intelligence to enable robots to identify, grasp, and place parts in real time, even when objects are randomly positioned, partially hidden, or when bins move during operation.

Conventional bin picking systems typically rely on fixed overhead 3D cameras, extensive calibration, and pre-defined grasp points, which can limit flexibility and increase system cost and downtime when conditions change. Inbolt’s approach places a 3D camera directly on the robot arm, allowing continuous perception and closed-loop control throughout the picking and placement process. According to the company, the system supports dynamic grasp selection rather than requiring a single predefined grasp point, enabling real-time adaptation during motion.

The bin picking workflow is modeled on human manipulation behavior. The robot identifies any accessible graspable surface, analyzes the object once it is in hand, and continuously refines its trajectory during motion to ensure accurate placement. This in-hand localization capability enables the system to respond immediately to part variability and bin movement, improving robustness compared with traditional vision-guided bin-picking approaches.

Inbolt reports that the solution achieves average processing times of under one second per pick and can reach success rates of up to 95 percent in live production environments. The on-arm camera configuration reduces hardware complexity by eliminating the need for multiple fixed cameras and overhead installations, allowing robots to be redeployed across different bins and workstations with reduced setup time.

“Traditional bin picking systems are too rigid for real factory conditions,” said Albane Dersy, COO of Inbolt. “We designed our solution to adapt in real time, able to see, grasp, and adjust the way a human would. That level of flexibility is what manufacturers need to reach truly autonomous production.”

The company states that the system is already deployed across more than five manufacturing facilities, where it has demonstrated consistent uptime and throughput in production environments. The platform runs on NVIDIA hardware and leverages Inbolt’s proprietary AI-based robot guidance models to perform real-time pose estimation and continuous trajectory correction while managing computational load across a range of part geometries. Manufacturers interested in deploying the system can request demonstrations or pilot installations through the company.

About Inbolt

Inbolt is a Paris-based robotics technology company specializing in real-time 3D vision and AI-driven guidance systems for industrial robots. The company develops on-arm vision solutions that enable robots to operate reliably in unstructured and dynamic manufacturing environments, supporting applications such as bin picking, part handling, and robotic manipulation across automotive, logistics, and industrial production settings. To learn more, please click here.

Source/Photo Credit: Inbolt

---

(Editor’s Note: All trademarks mentioned in this article, including company names, product names, and logos, are the property of their respective owners. Use of these trademarks is for informational purposes only and does not imply any endorsement.)