i-mation GmbH
Rottweil, Germany manufacturer-neutral machine-vision integrator founded 2001; builds 2D, 3D and deep-learning inspection and robot-guidance systems.
i-mation GmbH is an independent, manufacturer-neutral machine-vision integrator headquartered in Rottweil, Baden-Wurttemberg, Germany. Founded in 2001, the company has since built several thousand installed vision systems for industrial customers.
The company designs solutions across three vision-technology areas: classical 2D image processing, 3D image processing and structured-light scanning, and deep-learning-based vision. Because it is not tied to a single vendor, i-mation selects components across suppliers for each application. Named products include GampBOX, a verification system for pharmaceutical and medical packaging; i-ViTool, software for integrating deep-learning models into vision applications; and the VMB (Visual Machine Box) vision-computing hardware platform. Applications include code reading such as OCR, barcode and DataMatrix, completeness checks, positional-accuracy inspection, and vision-guided robot positioning for assembly and pick-and-place tasks.
i-mation serves customers in the automotive, solar, pharmaceutical, packaging, electrical-equipment, printing and robotics industries. Its quality system is certified to DIN EN ISO 9001, and the company holds AIA Certified System Integrator status with the Association for Advancing Automation. i-mation also belongs to the TechnologyMountains and MedicalMountains clusters in the Tuttlingen and Rottweil region.
Primary type & automation activities this supplier delivers:
GampBOX
Vision system for print/label inspection in medical and pharma packaging.
i-ViTool
Software tool integrating deep learning into Cognex VisionPro/Designer.
VMB (Vision Machine Box)
PC vision platform for high-end and deep-learning image processing.
Contact i-mation GmbH
WEBSITE
https://www.i-mation.dePHONE
+49 741 94 22 86-0HEADQUARTERS
Germany
Company Facts
Founded
2001
Primary Role
System Integrator
Company Size
-
Primary Region
Europe
Annual Sales
-
Funding Stage
-
Funding Total
-
Certifications & Memberships
Keywords
EPFL and NYU Develop Scalable Robotic Fish for Diverse Aquatic Environments
A team from EPFL and NYU has introduced ScaFi, a scalable robotic fish designed to adapt to various aquatic environments. Unlike traditional propeller-driven vehicles, ScaFi mimics the swimming motion of fish, allowing it to navigate shallow waters and dense vegetation without disturbing wildlife. The significance of ScaFi lies in its ability to maintain swimming efficiency across different sizes, which could reduce the engineering effort required for aquatic robots. The research highlights the importance of adaptable tools for environmental monitoring, as ecosystems vary greatly in size and complexity. Future developments will focus on understanding the energy efficiency of ScaFi as it scales, as initial tests revealed challenges with larger models. The team aims to explore whether the energy performance can be improved alongside the swimming capabilities of the robotic fish.
Chen Pu of Yuan Ke Vision Discusses Transitioning Robot Training Data from 2D to 4D with High Precision
On September 23, Chen Pu, Vice President of Product Development at Yuan Ke Vision, highlighted the importance of high-quality real data for robot training during a seminar in Beijing. He emphasized the need to elevate data from 2D to 4D and improve precision from centimeter to sub-millimeter levels to build a robust 4D data foundation for robotics. This transition is crucial as the industry faces challenges such as weak model generalization and inadequate scene adaptability. Chen noted that current training methods often rely on limited 2D video data, which fails to capture the complexities of three-dimensional space and temporal changes, hindering robots' ability to understand and interact with their environments effectively. Looking ahead, Yuan Ke Vision aims to address these challenges by developing new data collection and training paradigms that enhance dimensionality, enrich modalities, and improve precision. Chen pointed out that while video data serves as a foundation, it is often too simplistic, and the lack of tactile data remains a significant gap in the industry. No further timeline was disclosed at the time of publication.
Robot Training Data 4D Data Collection Industrial Robotics Embodied Intelligence Precision Robotics
Niantic Spatial Introduces Places Library Featuring 100 Real Environments for Robot Training
Niantic Spatial has launched its Places Library, providing robotics developers with a catalog of 100 real environments for training and evaluation. The assets, available as USDZ files, include two representations for each environment: a Gaussian splat for visual appearance and a mesh for collision detection. This initiative aims to enhance embodied AI training by offering realistic settings that align with gravity and support various simulators, including NVIDIA Isaac Sim. The significance of this launch lies in its potential to improve robot training efficiency. By utilizing real-world environments captured with a standard 360-degree camera, Niantic ensures that robots receive accurate visual and collision data, which is crucial for effective navigation and interaction. The library includes diverse settings, such as medical warehouses and urban streets, allowing developers to test their robots in varied scenarios without the need for extensive in-house data collection. Looking ahead, the Places Library could facilitate more advanced evaluations of robotic behavior in different environments. While the library's impact on performance across the 100 environments remains to be seen, it offers a valuable resource for robotics teams aiming to refine navigation tasks and adapt to changing surroundings. No further timeline was disclosed at the time of publication.
Niantic Spatial Flexion
Memo Unveils Physical-WAM to Enhance Embodied Intelligence at Global Digital Trade Expo
At the Global Digital Trade Expo on September 24, Memo introduced the Physical-WAM, the world's first physical-world action model equipped with physical perception capabilities. This innovation aims to address the challenges faced by robots in understanding and interacting with the physical world, which has hindered the widespread adoption of embodied intelligence. The significance of this development lies in its potential to bridge the data gap that currently limits robots' physical interactions. Memo's CEO, Li Minghao, highlighted that the lack of foundational physical understanding in robots stems from insufficient real-world interaction data and inadequate evaluation metrics. The Physical-WAM model is designed to enhance robots' predictive capabilities and real-time adjustments in physical tasks. Looking ahead, the demand for embodied intelligence is surging, with significant investments in the sector. However, challenges remain in aligning order fulfillment with funding growth. As the industry evolves, the effectiveness of solutions like Physical-WAM will be crucial in overcoming existing barriers to practical implementation.
Embodied Intelligence Physical AI Robotics Machine Learning
Robot Dogs Use Insect-Inspired Movements to See Through Obstructions in 17 Milliseconds
A collaborative team from Johannes Kepler University Linz, Free University of Bozen-Bolzano, University of Trento, and University of Sussex has developed a method for robot dogs to mimic insect movements to enhance their visual perception. By performing rhythmic lateral movements, akin to the 'peering' behavior observed in insects like locusts, these robots can effectively see through obstructions such as foliage and debris. This innovation is significant as it addresses a common challenge faced by exploration robots in environments cluttered with obstacles that hinder visual sensors. The research establishes a connection between the peering motion and synthetic aperture sensing, allowing the robots to create a virtual aperture that enhances background signals while blurring out obstructions. Future developments to watch include the potential applications of this technology in various environments, including forests and disaster sites. The research indicates that the synthetic aperture technique is most effective at around 50% obstruction density, suggesting a promising avenue for improving robotic vision in complex terrains. No further timeline was disclosed at the time of publication.
Robotics Computer Vision AI Obstacle Detection
Mondo Robotics' Beni: A Tumbling Robot Camera with Significant Obstacle Avoidance Issues
Mondo Robotics has introduced Beni, a small bipedal robot designed to capture and record video while following users. Equipped with a built-in camera and sensors, Beni can perform tricks like flips and roll while tracking specified objects via an app. However, after several weeks of testing, reviewers found Beni's obstacle avoidance capabilities severely lacking. The importance of Beni lies in its potential as a fun companion for personal activities like sports and pet video recording. Despite its interesting features, the robot struggles significantly with obstacle avoidance, often losing track of its target in cluttered environments. Reviewers noted that even in open spaces, Beni failed to navigate around obstacles effectively, which detracts from its overall usability. Moving forward, it will be crucial to monitor Mondo Robotics' efforts to improve Beni's obstacle avoidance technology. Currently priced at $599, the robot's value is questionable given its primary flaw. If Mondo can address these issues, Beni could become a more appealing option for users engaged in various activities, but until then, recommendations are limited.
Robot Cameras Obstacle Avoidance Consumer Robotics Video Technology
Noin Introduces GLOW Architecture for Instant Learning in Robotics
Noin has unveiled the GLOW architecture, a generative learning framework aimed at enhancing robot learning capabilities. This new technology allows robots to learn tasks instantly without the need for extensive data collection or predefined processes, addressing the challenges posed by dynamic real-world environments. GLOW's core innovation lies in its ability to enable robots to learn tasks through a single demonstration, transforming visual input and feedback into actionable tasks. The significance of GLOW is underscored by its potential to revolutionize how robots acquire skills, moving away from traditional methods that require separate strategies for each task. Instead, GLOW aims to equip robots with generalized capabilities that can be applied across various objects, environments, and tasks. This shift could lead to more efficient and adaptable robotic systems, as users will no longer need to break down their requirements into machine instructions. Future developments to watch include Noin's ongoing commitment to advancing the GLOW technology, which integrates visual understanding, spatial reasoning, task planning, and action generation within a unified model. The architecture has already shown promising results in various evaluations, indicating its effectiveness in physical scene understanding and task execution, paving the way for enhanced robotic intelligence and adaptability.
Robotics Generative Learning AI Embodied Intelligence
The Importance of a Comprehensive Network for Robotics: Insights from the 2026 Information Communication Expo
The 2026 China International Information Communication Expo highlighted the necessity of a comprehensive network for robotics, showcasing advancements in embodied intelligence. The expo featured technologies essential for robotic tasks, including perception, decision-making, motion control, and human-robot interaction, emphasizing the integration of these elements for effective operation. This focus on connectivity is crucial as robots must adapt to various environments, ensuring safety and efficiency in tasks ranging from industrial applications to healthcare. The expo demonstrated how robots can be integrated into smart factories and homes, where they must understand human needs and operate safely in shared spaces. The success of robotic systems relies not only on their ability to perform complex actions but also on their operational stability and cost-effectiveness. Looking ahead, the 2026 initiative for humanoid robots and embodied intelligence aims to enhance real-world training and application validation. The emphasis on reliable real-time control, stable connectivity, and effective data utilization will be key to advancing robotic capabilities and ensuring their successful deployment in diverse environments. No further timeline was disclosed at the time of publication.
Embodied Intelligence Robotics Technology Smart Factories Healthcare Robotics
Academician Zheng Nanning Discusses Embodied Intelligence Beyond Humanoid Robots
Academician Zheng Nanning, Director of the Artificial Intelligence and Robotics Institute at Xi'an Jiaotong University, published an article discussing the growth of embodied intelligence in the physical world. He emphasized that a key direction for artificial intelligence is enabling machines to not only process information but also to flexibly perform various tasks in dynamic environments. This includes precise operations on industrial production lines, autonomous navigation in complex terrains, and dynamic movements in exhibition settings. Zheng clarified that embodied intelligence is not synonymous with humanoid robots; rather, it focuses on whether intelligent systems can truly engage with real-world environments, forming a closed loop of perception, cognition, decision-making, action, and feedback. The evaluation of embodied intelligence should be based on its ability to solve real-world problems, not merely its physical form. He likened understanding embodied intelligence to an infant's process of learning about the world through interaction with their environment. To enable machines to act in the world, embodied intelligence requires various capabilities, including the formation of world models, multimodal representation, causal reasoning, and the integration of generation and action. The development of embodied intelligence relies on high-quality multimodal data and realistic simulation environments. Current limitations include the black box problem, computational constraints, and an immature industrial ecosystem. In sectors like industrial manufacturing, healthcare, and urban governance, intelligent systems are evolving from mere tools to proactive systems capable of continuous learning and adaptation.
Embodied Intelligence Artificial Intelligence Robotics Machine Learning
University of Washington Unveils 1-Gram Jumping Robot with Accurate Height Control
Engineers at the University of Washington have developed DirectHop, a 1-gram jumping robot capable of precise height control, enabling it to jump over standard stair steps. This innovation leverages a miniature motor and three folding legs, allowing the robot to perform multiple jumps with centimeter-level accuracy. The significance of DirectHop lies in its potential applications, which include detecting gas leaks in refineries, monitoring water and fertilizer usage on farms, and even exploring other planets. Unlike traditional jumping robots that rely on springs, DirectHop's design eliminates the need for complex mechanical components, making it easier to control jump distances and perform multiple jumps. Looking ahead, the team is exploring enhancements such as onboard solar cells, battery power, and navigation capabilities through cameras and sensors. Their ultimate goal is to achieve full autonomy, allowing DirectHop to navigate stairs independently by calculating jump heights and adjusting its position accordingly. No further timeline was disclosed at the time of publication.
Jumping Robots Micro Robotics Robotics Engineering Automation Technology
WAIC Review 4/4: From Port AGVs to Hospital Night Shifts: What WAIC 2026 Reveals About Real Robot Deployments
From 530+ AGVs at a Singapore port to robots on Apple's line, WAIC 2026 speakers shared concrete data on where embodied AI robots are actually working today.
Handbook of Marine Craft Hydrodynamics and Motion Control Thor I. Fossen — RSF Specialist Shelf · Book S1 of 6
Added mass, wave loading, DVL navigation: the marine GNC reference for USV, ROV and AUV service engineers. The hardest book on the RSF list, 5/5.
Reinforcement Learning: An Introduction Richard S. Sutton & Andrew G. Barto — RSF Specialist Shelf · Book S5 of 6
“It's the AI” will not satisfy an OEM support call. Sutton & Barto turn that into something a service engineer can investigate. Free PDF, Part I 4/5.
Planning Algorithms Steven M. LaValle — RSF Specialist Shelf · Book S4 of 6
The reference that treats planning as a subject in its own right: configuration space, sampling-based methods, RRT. Free online, dual difficulty 4/5-2/5.
Small Unmanned Aircraft: Theory and Practice Beard & McLain — RSF Specialist Shelf · Book S2 of 6
Beard & McLain derive the fixed-wing autopilot the RSF Top 10 never does: successive loop closure on roll, pitch, altitude and airspeed. 4/5.
Underactuated Robotics Russ Tedrake — RSF Specialist Shelf · Book S3 of 6
Legged robots fall over for control reasons, not mechanical ones. Tedrake's free MIT 6.832 notes explain why, and what a controller must do about it.
Robotics: Modelling, Planning and Control Siciliano, Sciavicco, Villani & Oriolo — RSF Specialist Shelf · Book S6 of 6
Siciliano's single-narrative counterpart to the Springer Handbook: kinematics through force control and visual servoing in one consistent notation. 4/5.
Probabilistic Robotics Thrun, Burgard & Fox — RSF Top 10 Robot Education Textbooks · No. 7
An AMR that lost localization broke nothing you can touch. Thrun explains Layer 4 faults: belief, Bayes filters, SLAM. Budget 12-15 hours.
Modern Robotics K. M. Lynch & F. C. Park — RSF Top 10 Robot Education Textbooks · No. 8
The book the field agreed to make free: full preprint PDF, video lectures and open code from the authors. Screw theory and kinematics, taught rigorously.
Introduction to Robotics J. J. Craig — RSF Top 10 Robot Education Textbooks · No. 6
Six of ten benchmark universities teach kinematics from Craig. Frames, transforms and D-H parameters: the language vendor manuals already speak.
