TechCapsules

Dynamic Creatures partners with Sixth Wall Studio to bring Hasbro characters to…

Dynamic Creatures announced a partnership with Sixth Wall Studio, Hasbro’s AI studio launched in June 2026, to create interactive robots that embody Hasbro characters. The collaboration will merge Dynamic Creatures’ mobile, expressive robots with Sixth Wall’s CharacterOS authoring and behavioral licensing, enabling characters to greet guests, remember them, and guide adventures in theme parks, resorts, hotels, cruise ships and retail venues. The first robot, Hugo, already demonstrates free movement and facial expression using the SnowJay platform’s 43 actuators. Both firms stress safety, initially keeping a human in the loop while gradually increasing autonomy. The joint effort aims to preserve each character’s voice and story while delivering personalized, memory‑enabled experiences for audiences aged 13 and older across worldwide entertainment ecosystems today.

💡 It turns beloved digital personalities into tangible, memory‑aware hosts, reshaping guest interaction in physical entertainment spaces.

Boston Dynamics adds wheels to Atlas humanoid to test whole-body control

Boston Dynamics released a video showing its Atlas humanoid equipped with wheels on its feet, used as a research testbed to explore the blend of deterministic physics‑based control and whole‑body reinforcement‑learning (RL) policies. Controls engineer Andy Wu attached off‑the‑shelf scooter rims to Atlas’ ankle actuators, allowing the robot to roll on flat terrain and switch to articulated legged motion for obstacles. The experiment, trained entirely in simulation, demonstrated real‑time proprioceptive feedback processing, dynamic pose and force adjustments, and rapid recovery from unmodeled disturbances. Wu’s goal was to push actuation limits, assess adaptability in unstructured environments, and study policy responses to actuator failures. The results suggest that combining classical motor control with whole‑body RL yields superior agility compared with either approach alone.

💡 Demonstrating seamless transition between wheels and legs proves a hybrid locomotion strategy can overcome terrain unpredictability, a key hurdle for deploying humanoid robots outside labs.

NVIDIA launches GR00T 2 and Isaac Lab 3.0 at CoRL

NVIDIA unveiled GR00T 2, an upgraded world‑action model, and Isaac Lab 3.0, the release of its robot‑learning framework, at the Conference on Robot Learning in Austin. GR00T 2 was trained on 100,000 hours of egocentric robot and human data and can complete 40 of 50 benchmark tasks without task‑specific demonstrations, achieving 71 % success with a single demonstration and 95 % with five. The model ranks in the top three across vision‑language‑action benchmarks and supports an API for sequencing skills and error recovery. Isaac Lab 3.0 adds environment setup, dual physics solvers (PhysX and Newton) and deformable‑object simulation, enabling parallel training of thousands of robots. Adopters such as Flexion, Siemens, Foxconn and university labs are already testing the tools for dexterous manipulation and industrial automation.

💡 By bridging the gap between simulated training and real‑world dexterity, GR00T 2 and Isaac Lab 3.0 give developers a faster path to deploy capable robots in manufacturing and research settings.

How a Mac mini powers a staggeringly capable robot without human training

Apple researchers have detailed a system that runs on an M4‑Pro Mac mini with 64 GB of RAM to control a Unitree G1 humanoid robot. The setup, named HALO (Humanoid Autonomy, Learned in Simulation, running Onboard), learns navigation and manipulation tasks in simulation before being deployed on the robot. In demonstrations, the robot autonomously maps unfamiliar rooms, crouches under tables, follows a person on command, and locates a chair when asked to sit. The paper, titled “Perceptive Humanoid Motion from Simulation,” includes video evidence of the robot analyzing environments and executing complex motions without human‑provided trajectories. Apple’s approach showcases the feasibility of using compact, consumer‑grade hardware for advanced robotic autonomy. The research highlights latency processing and efficiency achievable on the mini.

💡 By proving that a pocket‑sized Mac mini can drive a full‑size humanoid, Apple blurs the line between consumer computers and dedicated robotic controllers, opening pathways for rapid, cost‑effective deployment of intelligent machines.

Waymo robotaxis kept illegally passing school buses after software fix, feds say

Waymo’s self‑driving fleet has been under renewed federal scrutiny after the National Highway Traffic Safety Administration (NHTSA) opened a recall query into the company’s software fix for illegal school‑bus passing. The agency’s November 6 filing notes that most reported incidents occurred after Waymo deployed an update on November 17, 2025 intended to stop robotaxis from crossing stopped buses with flashing red lights. The investigation follows video from October 2025 showing a Waymo vehicle maneuvering around a school bus in Atlanta and multiple reports from Austin schools documenting similar behavior. Waymo asserts that improvements have eliminated citations since March 2026 and pledges to cooperate with NHTSA’s defect investigation. The probe also examines whether the algorithm’s learning updates were tested before rollout.

💡 If Waymo’s fix proves insufficient, regulators may demand broader software recalls, setting a precedent for how autonomous‑vehicle updates are validated.

How to minimize Wi-Fi roaming interruptions for autonomous mobile robots

Synaptics has unveiled a dedicated Wi‑Fi scanning architecture for autonomous mobile robots (AMRs) that separates network discovery from application traffic. The new Scan Core uses a low‑power subsystem to continuously monitor neighboring access points while the primary radio handles data, reducing roaming latency, packet loss, and power draw. AMRs, which may roam dozens of Wi‑Fi zones per shift, require sub‑80 ms latency and