Series: 'Events in Physical AI' (Oct 1–Oct 15) — Robots have left the lab and started real-world work
Series: "Events in Physical AI" (Oct 1–Oct 15) — Robots have left the lab and started real-world work
This article compiles news and developments in Physical AI from October 1 to 15, 2026, highlighting the transition of robots from lab settings to real-world applications. It discusses advancements in adaptive robotic systems, including improvements in spatial recognition, operational flexibility, and integration of AI with physical tasks. Key events include SoftBank's foundational model competition, AI-enabled transport robots, and demonstrations of human-like force application in robotics.
The article highlights the transition of Physical AI from controlled environments to real-world applications, emphasizing the shift from fixed-task automation to adaptive, context-aware robotic systems. It outlines key developments in Japan and overseas, focusing on the integration of AI with physical interaction, safety, and operational efficiency in logistics and service environments.
Source: Robotics — Robot learning news discovery · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
Model Development Competition" Domestic 2: AI recognizes misaligned pallets, improving the practicality of transport robots Announcement date: October 5, 2026 SoftBank Robotics announced that it would exhibit for the first time in the Kansai region a latest transport robot that uses AI to recognize pallets even if their placement is misaligned. [Commentary] In logistics workplaces, it is not always the case that car
Open source S11
AI" that learns human movements and force application Announcement/Reporting date: October 15, 2026 It was reported that H2L Inc. announced "BodySharing Physical AI," which utilizes proprioceptive data such as human posture, movement, and force application for learning. The report also introduced a figure of 95% work success rate when experience data is acquired for 15 minutes. [Commentary] Humans unconsciously adjus
Open source S15
ects empty containers Implementation date: October 15 and 16, 2026 In Marunouchi, Tokyo, a demonstration was planned where a robot equipped with Physical AI would hand drinks to people and collect empty containers placed in designated positions, and related news was reported on October 12. [Commentary] If it is just about grasping and moving objects, there are tasks that conventional robots can also do. However, to m
Open source S18
ensuring safety. In the future, when competing on robot performance, whether a robot can reliably adhere to what it must not do, in addition to what it can do, will become an important evaluation criterion. Source: NVIDIA Newsroom, October 13, 2026 Overseas 2: China's Hesai announces "Kosmo," a 3D camera supporting robot spatial awareness Report Date: October 14, 2026 China's Hesai Group has announced "Kosmo," a 3D s
Open source S21
patial mapping camera that combines LiDAR sensors and cameras to grasp surrounding space. [Commentary] Humans walk, grasp objects, and work while understanding the shape and position of objects and their distance from surroundings. For robots to work in similar environments, the ability to capture the three-dimensional structure of their surroundings is essential. Hesai has a track record in LiDAR for autonomous driv
Open source S22
Series: "Events in Physical AI" (Oct 1–Oct 15) — Robots have left the lab and started real-world work
Open source S1
Introduction AI is moving from the stage of writing text, creating images, and answering human questions to the stage of grasping, carrying, assembling, and working in cooperation with humans in the real world. What supports this change is "Physical AI." Following the news from the first half of October 2026, implementation in factories, logistics, and retail stores is progressing in Japan, while investment in AI bra
Open source S2
ins, spatial recognition technology, robot development platforms, and safety is expanding overseas. What is noteworthy is not just the novelty of the robots' appearance. To what extent have they come to understand the work that humans have been doing, and to what extent can they be entrusted with it? That is the key to deciphering the news this time. 1. Changes felt in events during this period From the news between
Open source S3
Source：Robotics — Robot learning news discovery · note.com