 
                                    Embedded transformer AI for light field delivery robot
 Cette publication existe aussi en Français
                                                Cette publication existe aussi en Français
                                            
                                                                            Vayu Robotics has developed a delivery robot to take on the systems from European supplier Starship Technologies.
Vayu has combined a transformer-based mobility foundation model with a light field camera sensor to eliminate the need for lidar. This light field, or plenoptic, sensing and transformer framework allows the Vayu One delivery robot to operate autonomously without pre-mapping the roads it intends to drive on and is capable of navigating inside stores, on city streets, and unloading packages on driveways or porches, carrying up to 40kg at under 20mph.
Vayu says the automatic loading mechanism (video below) makes the model the first-of-its-kind along with the lightfield camera (below) that takes in data from across the optical field and requires significant processing. The company also spoke at Nvidia’s GTC conference in March on the technology, implying that the transformer framework runs on a Jetson AI card.
Vayu was founded by three executives from the robotics and mobility industry, Anand Gopalan, former CEO of lidar supplier Velodyne, Mahesh Krishnamurthi, formerly Apple SPG and Lyft, and Nitish Srivastava, also from Apple SPG and Geoffrey Hinton’s AI lab in the University of Toronto. Geoffrey Hinton, often described as the ‘godfather’ of AI, is also an advisor to the company.

“The unique set of technologies we have developed at Vayu have allowed us to solve problems that have plagued delivery robots over the past decade, and finally create a solution that can actually be deployed at scale and enable the cheap transport of goods everywhere” says Gopalan, CEO of Vayu Robotics.
“Our software is robot form factor agnostic and we have already deployed it across several wheeled form factors.In the near future, Vayu’s software technology will enable the movement of quadrupedal and bipedal robots, allowing us to expand into those markets as well,” said Gopalan.
Vayu’s Delivery Robots are already being deployed, with a contract with a large e-commerce player to deploy 2500 robots. The team is also working with a leading global robotics manufacturer to replace lidar sensors with Vayu’s sensing technology for other robotic applications.
Vayu has previously raised $12.7 million from leading investors Khosla Ventures and aerospace firm Lockheed Martin.
“At Khosla Ventures, we believe in backing businesses where critical and differentiated technologies can unlock a large market. Vayu is a great example of this where they have deployed novel sensing and their AI foundation models to a robotic challenge that can have immense economic and societal impact” said Kanu Gulati, Partner at Khosla Ventures.
“Autonomous delivery robots are only the tip of the iceberg,” said Anand Gopalan. With its cutting-edge innovation and deployment, Vayu is poised to lead the adoption of real-world robotics across industries. For now, Vayu’s scalable robotics architecture is set to empower small businesses to deliver products to their customers’ doorstep seamlessly.
 If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
                                        
                                            
                                               eeNews on Google News
                                        If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
                                        
                                            
                                               eeNews on Google News
                                        
                                                                     
                    
                 
                    
                 
                    
                 
                    
                 
                    
                 
                                            
                                        