Fresh developments from this year’s Automate conference have surfaced, courtesy of Alphabet X spinoff Intrinsic. Revealed at the event in Chicago, the company disclosed its integration of several Nvidia solutions into its Flowstate robotic app platform.
Among these incorporations is Isaac Manipulator, an array of foundational models tailored for crafting workflows for robot arms. Initially introduced at GTC in March, the offering has already garnered support from major players in industrial automation such as Yaskawa, Solomon, PickNik Robotics, and more.
The collaboration is primarily honed in on grasping, a critical aspect for manufacturing and fulfillment automation. These systems, trained on extensive datasets, aim to execute tasks that transcend hardware limitations, fostering adaptability across diverse scenarios.
Intrinsic’s founder and CEO, Wendy Tan White, emphasized the future potential of universal grasping skills in expediting programming processes. Early trials of the Flowstate platform took place in Isaac Sim, Nvidia’s robotic simulation environment, with Intrinsic’s customer, Trumpf Machine Tools, actively engaged in prototype testing.
Showcasing Intrinsic work with Google DeepMind
Moreover, Intrinsic is teaming up with DeepMind, another Alphabet-owned entity, to tackle pose estimation and path planning, essential components of automation. Leveraging machine learning, their collaborative efforts have yielded promising results, including the ability to operate multiple robots simultaneously.
- Multi-robot motion planning with machine learning: Today, developers often spend anywhere from a few hours on simple applications to hundreds and even thousands of hours on complex multi-robot applications. With Intrinsic’s and Google DeepMind’s efforts on a universal AI-based robot motion planner for one or multiple robots sharing the same workspace, this work can be done automatically.
- Learning from demonstration, applied to two-handed dexterous manipulation: To create scalable and generalizable solutions, many assembly applications require robots to manipulate objects with two arms and grippers. The ability to gently pick up, move, precisely place, and assemble objects with human-like dexterity is highly challenging and valuable for robotics at large.
Collaborating with DeepMind and understanding each other’s needs has helped validate the decision to fully integrate AI into the Intrinsic platform.
The prospect of employing dual-arm systems, akin to humanoid robots, also looms large on Intrinsic’s agenda. This shift holds significant implications for the broader landscape of robotics, promising expanded capabilities and applications in the near future.