An adaptive robot seamlessly fuses industrial-grade force control with advanced artificial intelligence. It can adapt to complex environments and automate various tasks in much the same way a human does.
Primary use case: Force-controlled automation with position-error tolerance, disturbance rejection, and human-like hand-eye coordination.
Locomotion
unknown
Hand Type
unknown
Autonomy Level
unknown
AI Capabilities
vision, manipulation
Availability
unknown
Primary Use Case
Force-controlled automation with position-error tolerance, disturbance rejection, and human-like hand-eye coordination.
Boston Dynamics unveiled a new four-finger hand for its Atlas humanoid robot, featuring 13 degrees of freedom including a four-DOF opposable thumb and dense tactile pressure sensors. The design was optimized for complex manipulation tasks such as in-hand object reorientation, slip recovery, and tool operation including drills, grinders, and welding torches. The company deliberately chose four fingers over a five-finger human-like hand to reduce cost, size, and failure points while maintaining the dexterity needed for industrial work.
Nvidia has launched Isaac ROS 5.0, introducing AI agent capabilities designed to help developers build, customize and deploy robotics applications faster within the open-source ROS ecosystem. The release, announced at ROSCon in Toronto, adds agentic development workflows and expanded support for Nvidia's Jetson platform and physical AI technologies.
NVIDIA released Isaac ROS 5.0 at ROSCon 2026 in Toronto, introducing AI agents to robotics development to help automate repetitive tasks, streamline setup and manipulation workflows, and reduce friction in the development process. The release includes support for ROS 2 Lyrical and Ubuntu 24.04, FoundationPose for 5.5x faster object pose estimation, and agent-ready skills for pick-and-place workflows.