Accelerating the robotics revolution with AI


AI is making robots easier to use whilst enabling them to take on more challenging jobs. As the capabilities of the technology continue to expand, it will create the momentum for a surge in robotic automation, predicts Julian Ware, UK & Ireland Sales Manager for ABB Robotics. 

Industrial robotic systems are already highly adaptable and capable, contributing to their rapid growth across the globe, but with artificial intelligence (AI), this growth looks set to move to a whole new level, with recent estimates from global research firm Statista predicting a 13.5% increase in the AI industrial robotics market, from US$8.49bn in 2023 to US$20.64bn by 2030.

Although robots bring flexibility, resilience and sustainability and help meet changing customer demands while addressing labour and skills shortages, many UK companies have yet to embrace robotic automation. In our recent survey of 250 UK industrial companies, 59% of respondents currently do not use robots, citing lack of experience and perceived complexity as their key reasons for not investing in robotic solutions.

Developments in AI are steadily sweeping this away by making robots much easier to train and use. AI-enabled robots can see, act and reason in the world around them. Using reinforcement learning, these robots can adapt to new tasks and help open new opportunities and applications over and above those available with conventional programmed robots.

These developments, together with tools such as easy programming and simulation software and controllers offering enhanced functionality, are helping to remove many of the skills barriers that have hampered the adoption of robots. As such, we can expect to see robots employed in far greater numbers outside of their traditional roles in manufacturing and distribution, with companies in industries including electronics, healthcare, e-commerce, pharmaceuticals, and food service already finding ways to incorporate robotic automation into their operations.

All-seeing AI

Many AI and vision-based robotic installations are related to sorting and picking. For example, ABB’s Robotic Item Picker features a robot, suction grippers, and a proprietary machine vision sensor to automate complex picking tasks for cuboids, cylinders, pouches, boxes, polybags, and blister packs, among others. Combining machine vision and artificial intelligence, the Item Picker determines the optimal grasp points for each item before the suction gripper picks up and places the item into designated bins.

The system doesn’t need any human supervision or information about the physical attributes of the items it picks. With a picking rate of up to 1,400 items per hour, the Item Picker enables businesses to handle more orders without increasing headcount or time, directly addressing the labour shortage challenges increasingly facing businesses.

Another use of machine vision is in Autonomous Mobile Robots (AMRs), which can move independently around a facility. ABB Robotics is transforming its AMRs by adding Visual Simultaneous Localisation and Mapping (Visual SLAM) technology that enables them to make intelligent navigation decisions based on their surroundings.

Visual SLAM combines AI and 3D vision technologies, with cameras mounted on the AMR to create a real-time 3D map of all objects in the surrounding area. The system can differentiate between fixed features such as floors, ceilings, walls, and objects such as people or vehicles that can move or change position. The cameras detect and track features in the environment, enabling the AMR to dynamically determine the safest and most efficient route to its destination.

Visual SLAM has several benefits over alternative AMR navigation systems, with inflexible forms of navigation such as magnetic tape, QR codes and traditional 2D SLAM all requiring additional infrastructure to function. Unlike 2D SLAM, Visual SLAM requires no additional references such as reflectors or markers. This saves cost and space and offers accurate positioning to within three millimetres.

With no need to stop production or add infrastructure such as reference points, Visual SLAM can reduce commissioning time by up to 20% compared to 2D SLAM and can also be deployed at scale by updating entire AMR fleets remotely.

Despite their added freedom, this additional autonomy and intelligence allows new AMRs to operate just as safely as their predecessors in changeable environments featuring human workers.

Thanks to AI’s ability to make robots easier to use, integrate, and access, the use of AI and autonomous technologies at scale will be a major trend for robotics, with new applications unfolding as new capabilities are added

By making established applications easier to program, operate and maintain, autonomous technologies should prove instrumental in helping many more companies to make their first investments in robots or to use them in new applications.


About Author

Comments are closed.