In a LinkedIn post accompanied by a captivating video, Sanjeev Sharma, the Founder and CEO of Swaayatt Robots, has unveiled remarkable progress in the realm of autonomous driving. The video presentation showcased Swaayatt Robots’ autonomous vehicle navigating through challenging off-road terrain, confronting both static and dynamic obstacles with an astonishingly low margin of error.
The company utilized very sparse stochastic HD maps for stochastic projection of boundary information in the presented demonstration. One of the most noteworthy aspects of this demonstration was the vehicle’s ability to negotiate bi-directional traffic on a single-lane off-road path, thanks to novel motion planning and decision-making frameworks.
According to the CEO, this remarkable achievement demonstrates the company’s commitment to creating cutting-edge technology for autonomous vehicles. The Indian-based company, Swaayatt Robots, is renowned for its pioneering work in the field of robotics and autonomous vehicles, pushing the boundaries of what’s achievable in this dynamic industry.
According to the post, the company plans to further enhance this capability by incorporating multi-agent intent analysis and bi-directional negotiation policies developed through deep unsupervised learning and reinforcement learning techniques. As the company takes its initial steps towards this grand vision, they will soon execute a similar high-speed demo in a single-lane on-road environment.
Moreover, Swaayatt Robots is actively working on eliminating the reliance on high-definition maps for autonomous driving in off-road conditions. They aim to have their autonomous vehicles build cost-maps and understand environmental boundaries using GPS nodes as waypoints, driven by unsupervised deep learning.