PRESS RELEASE – Duality AI, the corporate behind Falcon, a digital twin simulation platform, at present introduced an settlement with NASA’s Jet Propulsion Laboratory (NASA JPL) in Pasadena to proceed its work on Protection Superior Analysis Tasks Company’s (DARPA’s) Robotic Autonomy in Complicated Environments with Resiliency program (RACER). This announcement follows Part I of the RACER challenge, the place Duality efficiently accomplished this system, as one among solely two simulation suppliers chosen by DARPA. The settlement accelerates autonomy growth for NASA JPL by permitting speedy iteration and discovery of flaws via simulation, decreasing dangers and offering the artificial information very important to coaching superior AI algorithms that push the boundaries of robotic autonomy.
“Simulation know-how allows a sensible manner for NASA JPL to coach their techniques in digital environments and generate artificial information that’s important for highly-capable AI fashions to be prepared for any problem.”
Put up this
In the course of the second section of the RACER program, groups should show their skill to create superior AI options that permit large-scale automobiles to journey autonomously at sooner speeds and over longer distances than in Part 1. The group from NASA JPL will use Duality’s superior digital twin simulator, Falcon, to beat the challenges related to autonomous high-speed, off-road automobiles, together with mapping, notion, planning, and responding to hazards and obstacles. The ultra-realistic artificial information generated by Falcon is pivotal within the coaching and testing of the next-generation AI techniques.
“Success within the RACER program depends on coaching clever techniques to finish complicated duties that conventional autonomous automobiles don’t face. From sudden obstacles to unpredictable climate challenges, the variety of eventualities these AI techniques should be educated and examined towards is astounding,” mentioned Mike Taylor, Chief Product Officer and co-founder of Duality AI. “Simulation know-how allows a sensible manner for NASA JPL to coach their techniques in digital environments and generate artificial information that’s important for highly-capable AI fashions to be prepared for any problem.”
As a part of the settlement, Duality will present the instruments and processes required to coach AI via the usage of digital twin simulation and support within the growth of NASA JPL’s subsequent technology of autonomy applied sciences. The contract additionally offers NASA JPL with entry to expansive, extremely reasonable digital twin environments crucial for sturdy testing and coaching of autonomy software program. The digital twins of actual areas allow the NASA JPL group to generate related artificial information and take a look at their autonomy software program with the next share of code protection than attainable with standard simulators. By leveraging Duality’s AI-powered instruments to construct digital twin environments from satellite tv for pc imagery, as nicely their library of configurable digital sensors, NASA JPL can simply manipulate digital twin environments with completely different rocks, bushes, climate, and extra. This degree of exact management helps NASA JPL transition cutting-edge concepts from simulation into the true world.
Duality’s contract with NASA JPL advances Duality’s mission to construct a digital twin simulation platform that helps clear up essentially the most complicated robotics issues. Working alongside high minds in robotics permits Duality to boost its simulation capabilities and show the ability of its know-how by overcoming immense real-world challenges.
To be taught extra about Duality and its suite of Falcon merchandise for digital twin simulation go to duality.ai.