Oxbotica is a U.K.-based company that takes care of the navigation, perception, user interface, fleet management, and other intricate features and software required to power self-driving vehicles. This startup was launched in 2014 by Paul Newman and Igmar Posner, both professors at the Oxford University. From then onwards, this company has grown from one strength to another. The main aim of Oxbotica is to create software that drives autonomous vehicles without having to depend on any third-party for the conduction of operations. Oxbatica recently raised $47 million of funding in the Series B round, taking the net funding to about $76.4 million. This funding would be primarily utilized towards accelerating the commercial deployment of the software in the leading industries and markets all over the globe.
Self-driven cars have captured millions of people’s attention, but these cars have not been deployed on a large scale due to safety concerns. Oxbotica aims to bridge this gap and provide software that acknowledges all the situations. The basic technology in play here is a general adversarial network (GAN) based simulation technology-selenium. The GAN works as a generative model that employs two neural network models where one model is known as the generator, and the other is the generative network. This technology, when applied successfully, generates photorealistic images within a brief time.
The technology can reverse the road signage and even change a particular image’s lighting to demonstrate the different times of the day or even the seasonal change. For example- it can replace a tree with a building in the photorealistic image while also ensuring that the reflection indicates precisely what it should. These synthetic images are deployed to teach the software aptly and to verse it well with reality. The technology generates every possible experience down to even raindrops.
The platform that Oxbotica is developing seeks to include mapping, simulation, indexing, logging, diagnostics, and vehicle to vehicle communication. This software will also need less computing power and would easily blend in with cloud management. Safety will be the primary concern in the development of this software. For that purpose, lidar sensors, radar for localization and perception have been put into place. All these will increase the fleet’s intelligence, and therefore the safety of any individual will not be put into jeopardy by the self-driven vehicle. This software has been practically implemented on private roads, test tracks, and simulations for hundreds of thousands of miles. It has been found to work in almost all scenarios across the American, European, and Australian tracks. Oxbotica also plans to retain the vehicle’s manual control, thereby making the vehicle function dually. The company also conducts regular penetration testing to look out for potential hackers who could wreak havoc.
With the world facing a pandemic, this technology’s deployment is likely to take place sooner than expected. Self-driving cars, vans, and trucks will help control the pandemic as it will reduce person-to-person contact. This could also potentially assist in establishing the trust of people with autonomous vehicles.
Amreen Bawa is a consulting intern at MarktechPost. Along with pursuing BA Hons in Social Sciences from Panjab University, Chandigarh, she is also a keen learner and writer, having special interest in the application and scope of artificial intelligence in various facets of life.