Dan Price ONPASSIVE NZ 😃’s Post

The latest research on 'Direct Comparisons of Upper-limb Motor Learning Performance among Three Types of Haptic Guidance with Non-Assisted Condition in Spiral Drawing Task' is now published in IEEE Transactions on Neural Systems and Rehabilitation Engineering. Delve into these findings and discover the impact of different haptic guidance types on motor learning. Check out the full paper here: https://bit.ly/3WyqqfE Additional video: https://lnkd.in/eCgNNweX Authors: Hisayoshi Muramatsu, Yoshihiro Itaguchi, Chiharu Yamada, and Hiroshi Yoshizawa. TNSRE Editor-in-Chief: He (Helen) Huang #robotics #rehabilitation #research #hapticguidance

Our latest work, accepted for #ICRA2024, on Physics-informed Neural Networks for Constrained Robot Motion Planning, is now available online. Unlike prior learning-based methods, this approach doesn't require expert trajectories for training. Instead, it directly learns to solve Eikonal PDE on constrained manifolds, resulting in data generation times of just a few seconds. Moreover, it demonstrates faster planning times than previous methods, even in high-dimensional complex scenarios and real-world object manipulation tasks. The following demo showcases our method for solving tasks such as door opening and moving a cup without tilting between confined locations. Paper: https://lnkd.in/deVABr28 #purdueresearch #purduecs #robotics #machinelearning #PINNs

Excellent insight!

Federated learning has a huge potential to help train at the edge for senarios where the training data has privacy concerns, especially for robot operations with cameras. Robot Foundation Models will disrupt the robotics space in my opinion not just for consumers but also for developers. There is also a huge potential for monetisation where specific skills for a robot will be trained with robotics Labs and offered as “SaaS” for any kind of robot that is compatible. Difference is it’s not the traditional software that will be shipped for on board runtime but a weight corresponding to a skill to do X. The future is already here!! OctaiPipe #robotics #Covariant #figureai

Check out the innovative P1 biped robot by LimX Dynamics! Designed for stable walking in complex environments using advanced reinforcement learning, P1 features high-performance joint modules, real-time communication, and user-friendly software. With a Real2Sim2Real closed-loop system, robust neural network architecture, and a range of sensors, it's paving the way for efficient simulation-to-hardware deployment. Specs include a 48V battery, 1000W power, and up to 2 hours of operation. Follow this page for the latest development in AI #topaitools #MachineLearning #Robotics #AI #Innovation #Technology

New AI Robot with Human Brain Shocks the World! (They've Crossed the Line) Recent advancements in AI and robotics have led to significant breakthroughs, including a robot with a brain-on-a-chip in China and Skild AI's development of a universal robot brain for complex tasks. Meanwhile, Zen Technologies in India has introduced Prahasta, a four-legged robot equipped with advanced LIDAR and AI for defense purposes, capable of navigating difficult terrains and carrying heavy loads. These innovations highlight the rapid evolution of robotics technology, blending artificial intelligence with physical capabilities in unprecedented ways, poised to transform industrial, defense, and healthcare sectors. https://lnkd.in/dfFTX6nJ

I promise, that's my last post re. Covariant's RFM-1. But, the implications of this new approach are so big because this type of software will build a foundation for a new approach to teach and operate robotic applications. And coming from a financial point of view, this software will also enable the build up of robotic-based asset management portfolios. #AI #robotic #asaservice #EaaS

New AI Robot with Human Brain Shocks the World! (They've Crossed the Line) Recent advancements in AI and robotics have led to significant breakthroughs, including a robot with a brain-on-a-chip in China and Skild AI's development of a universal robot brain for complex tasks. Meanwhile, Zen Technologies in India has introduced Prahasta, a four-legged robot equipped with advanced LIDAR and AI for defense purposes, capable of navigating difficult terrains and carrying heavy loads. These innovations highlight the rapid evolution of robotics technology, blending artificial intelligence with physical capabilities in unprecedented ways, poised to transform industrial, defense, and healthcare sectors. https://lnkd.in/drPqDJXN

TossingBot: Learning to Throw Arbitrary Objects Researchers have trained a robot to throw objects using visual observations and trial and error, overcoming challenges posed by varied object poses and properties. By combining simulation and deep learning, the system achieves accuracy and generalization for new objects and target locations. This technology might find applications in sorting rubbish or kitting non-sensitive parts, allowing robots to rapidly place objects beyond their kinematic range in pick-and-place scenarios. Thoughts ⁉️ #robotics #simulation #gripper #research #cobots Via: Andy Zeng from Princeton University Inspired by: Daniel Kuepper

TossingBot: Learning to Throw Arbitrary Objects Researchers have trained a robot to throw objects using visual observations and trial and error, overcoming challenges posed by varied object poses and properties. By combining simulation and deep learning, the system achieves accuracy and generalization for new objects and target locations. This technology might find applications in sorting rubbish or kitting non-sensitive parts, allowing robots to rapidly place objects beyond their kinematic range in pick-and-place scenarios. Thoughts ⁉️ #robotics #simulation #gripper #research #cobots Via: Andy Zeng from Princeton University Inspired by: Daniel Kuepper