Robot arms have been used for decades, but most success has come in stable, high-volume settings such as automotive manufacturing. Coleman says new applications face changing objects, surroundings, and lighting, and explains that traditional machines can be more practical than arms for simple, repeatable jobs.
Coleman compares bespoke robotics to the gap between large hotels and short-term home stays: many teams need something between turnkey industrial automation and building an entire system themselves. Industrial tools are cost-effective and reliable for predictable tasks, while custom stacks offer flexibility but demand substantial research and development.
Coleman describes a robotics stack that combines machine learning with established tools such as collision checking, motion planning, inverse kinematics, ROS, and simulation. He recommends behavior trees for organizing actions and recovery, and emphasizes that teams need extensible code so they can add application-specific logic.
Robot arms can plan around scanned environments and moving obstacles instead of relying only on fixed waypoints. Coleman highlights force control, point-cloud processing, and machine learning for tasks such as segmentation, grasp generation, and pose estimation, with conventional safety checks supervising learned outputs. Simulation lets teams develop and test software before the robot hardware is ready.
Coleman argues for an open, modular developer platform that supports different robot arms, cameras, sensors, and visualization tools, with commercial support. PickNik's MoveIt Pro is presented as one such platform, built around ROS and intended to help teams develop applications faster and at lower cost. He also describes growing work with mobile robot arms and multi-arm systems, while expressing skepticism about the cost and complexity of bipedal humanoids.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Jason Osborn of Intel Robotics talks with Dave Coleman, PickNik Robotics founder and chief product officer, at Automate 2026 in Chicago about robot control, compute choices, safety, and a shelf-stocking demo.
Dave Coleman, founder and chief product officer at PickNik Robotics, discusses robotics software and the UX of MoveIt Pro at a Northwest Robotics Alliance talk.