Swaayatt Robots (स्वायत्त रोबोट्स)

As we further our Level-5 autonomous driving research, and scale up the technology, including sustainable last-mile autonomy operational capabilities, we begin our day-time operations on regular urban roads in India. This is to ensure that our #autonomousdriving technology can work globally and can negotiate any kind of environments and traffic-dynamics. We will gradually increase the complexity of the traffic-dynamics and the environments that our #autonomousvehicles are tasked to negotiate. In this demo, the vehicle was tasked with navigating a regular 4-lane city road, with end goal being able to park itself in a very constrained unstructured parking spot. Often, even on 4-lane roads, vehicles, bicyclists, and autos can come from the wrong side. We mimic this stochastic traffic pattern by having motorcyclists come from the wrong side and not yield to our vehicle -- leaving it to our vehicle to decide whether to come to a standstill or to yield from its current driving lane and perform an avoidance maneuver. At the end, we demonstrate last-mile autonomy capability by having our autonomous vehicle park itself in a very unstructured parking spot, with bikes parked randomly in the area, outside of our office. We had earlier demonstrated such capabilities in structured campus environments in our demos in January and February 2023. As can be seen from the demo, vehicle executed these tasks in a generic open environment, while also negotiating a traffic-intersection, and crossing the divider on the 4-lanes road. This is only a beginning of our day-time operations on roads in India. #reinforcementlearning #deeplearning #machinelearning Swaayatt Robots (स्वायत्त रोबोट्स)

Projective 3-Point (P3P) algorithm plays a crucial role in understanding the Visual Odometry and typical SLAM algorithmic pipelines. P3P algorithm allows us to compute camera's (robot's) transformation using the 3D-2D correspondences of feature points in a camera's 2D image plane and their corresponding 3D real-world coordinates. P3P algorithm requires 3 such feature points in an image and their corresponding 3D coordinate. The previous set of lectures discussed the mathematical formulation of the problem statement and step by step solution. The geometry involved, via P3P algorithm, in computation of the camera transformation parameters results in a quartic equation, resulting in 4 possible solutions. This lecture graphically discusses the four possible solutions and how we can use a 4th feature point in an image to get the true transformation. This lecture is a part of my course CV-2.0X: Introduction to Visual Mapping and Odometry at Deep Eigen. We are also providing courses in #reinforcementlearning #computervision #machinelearning and #deeplearning. Upcoming courses also include advanced machine learning with modern deep learning concepts, and mathematical optimization.

On this Independence Day, at Swaayatt Robots (स्वायत्त रोबोट्स) we have taken a pledge of bringing autonomy to all frontiers, i.e., Civil, Military, and Industrial, both on the ground and in the air. Over the years, specifically in the past 9 months, we have demonstrated our #autonomousdriving technology conquering on-roads, off-roads, and geo-fenced autonomous navigation. We are going to hereafter scale it up for multitude of applications. Wishing everyone a very Happy Independence Day! Jai Shree Ram 🙏 Jai Maa Kali 🙏 Har-Har Mahadev 🙏 Jai Hind! 🇮🇳 #autonomousvehicles

We are pleased to announce that we have raised $4M, as a part of our 2nd seed round, which is part of a large fund, that we are raising with global investors. Going forward, we will ever more actively be inventing new AI capabilities to solve the problem autonomous general navigation, and scale up #autonomousdriving to new heights. This scaleup will begin with significant capability enhancements in our existing models like: (1) Toll-Plaza navigation framework, (2) Bidirectional traffic negotiation on single lane roads framework -- enabling relative speeds exceeding 65 KM/H at points of intersection of autonomous agents, (3) Generic traffic flow negotiation off-roads, (4) Unmanaged traffic intersection negotiation, (5) Multi-Agent Intent Analysis and Negotiation, and adding strong some new capabilities such as (and many more): (1) Learning to overtake on typical two lane roads akin to Indian drivers, (2) Developing new inverse-#reinforcementlearning paradigms for abstraction learning for both on- and off-roads autonomous navigation, (3) Redundancy of perception algorithms for off-roads navigation. A major objective includes demonstrating tech-architecture for scalable, safe, and compute efficient L4 #autonomousvehicles, fuelled by our L5 R&D and algorithmic frameworks. Our foundational belief is: Unless we solve the hardest AI problem, i.e., autonomous agents learning to negotiate the most (i) complex, (ii) stochastic, and (iii) adversarial traffic-dynamics on the roads, to control the behaviour of a nonlinear dynamical system operating in a continuous domain, such as an autonomous vehicle, it is impossible to guarantee safety and operationally cost effeciency. With limited resources, the way we approached this is: We picked the most complex frontiers in this domain, and developed AI/mathematical models and algorithms to solve those problems, like: (i) bidirectional traffic negotiation (relative speeds at point of intersection 40 KMPH) on single lane roads, (ii) negotiating multi-agents in tight stochastic dynamic environments, (iii) toll-plaza, (iv) off-roads navigation at high-speeds along with negotiating bidirectional flow of traffic, (v) unmanaged traffic-intersection negotiation, (vi) LiDAR-less perception (day / night), etc. The outcome of all of this work is safe, robust, and cost effective autonomous general navigation technology that is enabling us to do what no one else has done before in this industry: doing both on- and off-roads generic navigation simultaneously -- typically considered as different verticals. There is much more to come going forward, and this is only a beginning. #deeplearning #machinelearning

Bhopal-based AI and robotics startup Swaayatt Robots (स्वायत्त रोबोट्स) has raised $4 million (approx. Rs 33.4 crore) from American investors to make autonomous cars a reality in India. With this funding, the startup is now valued at $151 million. Interestingly, Swaayatt Robots plans to further raise $11 million at a valuation of around $175-$200 million. "Going forward, with this new fundraise, we will actively be inventing new AI capabilities to solve the problem of autonomous general navigation," Sanjeev Sharma wrote in a LinkedIn post. He said that the startup aims to solve the hardest AI problem to ensure safe and affordable autonomous driving technology. Founded in 2015 with an initial investment of Rs 80 lakh, Swaayatt Robots has developed exclusive algorithms and claims that it can adeptly navigate busy Indian roadways without human assistance. According to the Society of Automotive Engineers International, autonomous driving is categorized into six levels, ranging from Level 0 to Level 5. Swaayatt aims to attain Level 5 autonomous driving. The startup has so far carried out more than 80 demos, although it has not yet disclosed its completely autonomous capabilities on public roads. #bhopal #startup #cars #autonomous #ai #investors #indianstartupnews

Bhopal-based autonomous driving startup Swaayatt Robots (स्वायत्त रोबोट्स) has secured $4 million in funding from undisclosed US-based angel investors. This latest investment round values the company at $151 million. According to Swaayatt Robots' visionary founder, Sanjeev Sharma, the $4 million raised is part of a larger funding round. The startup aims to raise additional funds at a valuation of around $175 million over the next six to seven months. The fresh capital will be utilized to expand research and development efforts in critical areas such as reinforcement learning and unsupervised learning. The ultimate goal is to develop and demonstrate a sustainably scalable Level 5 autonomous driving technology architecture. Level 5 autonomy, as defined by the Society of Automotive Engineers (SAE), represents the highest level of self-driving capability, where the autonomous system can handle all driving functions in all conditions without any human intervention. Sharma stated on LinkedIn about Swaayatt being a pioneer in #AI and #autonomousdriving: “…we were a first mover of #reinforcementlearning applied to a real-world nonlinear dynamical system operating in a continuous domain, an autonomous vehicle, using 5 autonomous (RL-) agents. This was the time when Deep Mind was the only company working at that scale in RL (AlphaGo, AlphaZero).” Since its inception in 2015, Swaayatt Robots has developed proprietary algorithms and claims to have unparalleled capabilities in autonomous navigation, even on the typically chaotic Indian roads. During its self-funded phase from 2016 to 2021, the startup demonstrated robust perception algorithmic frameworks, including pioneering LiDAR-less perception and ensuring night operations.

Bhopal-based autonomous driving startup Swaayatt Robots (स्वायत्त रोबोट्स) has announced the raising of $4 million in funding from US-based angel investors. The founder of Swaayatt Robots, Sanjeev Sharma, announced on LinkedIn that the funding is part of a much larger round the startup is raising, with additional funds expected at a $175 million valuation over the next six to seven months. Previously, in 2021, Swaayatt Robots had raised about $3 million. Sharma also stated in the post that the startup needs to find a solution to the hardest AI problem to ensure safe and affordable autonomous driving technology. Since its launch in 2015, Swaayatt Robots has created exclusive algorithms and asserts that it can adeptly navigate busy Indian roadways without the need for human assistance. Follow Rise & Shine for daily startup updates. Stay informed, stay ahead ➡️ www.riseshine.in #autonomous #funding #ai #startup #funding #linkedin

🔥Swaayatt Robots (स्वायत्त रोबोट्स) 🔥Sanjeev Sharma 🚀🔥Swaayatt Robots Raises $4 Mn To Make Autonomous Cars A Reality🚀🔥 🌟Driving Innovation: Swaayatt Robots' $4 million raise marks a significant milestone in making autonomous cars a reality, paving the way for a smarter, safer future. 🌟Pioneering Autonomy: With this funding, Swaayatt Robots is set to revolutionize transportation, leading the charge in autonomous vehicle technology. 🌟Visionary Leadership: The success of Swaayatt Robots highlights the power of visionary thinking and relentless pursuit of groundbreaking advancements. 🌟Future of Mobility: This investment brings us closer to a world where autonomous cars enhance efficiency, reduce accidents, and transform our mobility experience. ✅️Share Your Startup Story 🚀📲 https://lnkd.in/ggir3Qap ✅️visite our website🌐 https://lnkd.in/dad4nHP4 #AutonomousVehicles #TechInnovation #FutureOfMobility #SwaayattRobots #scoopearth

Swaayatt Robots (स्वायत्त रोबोट्स) Robots, an autonomous driving startup based in Bhopal, has secured $4 million in funding from US-based angel investors. Founder Sanjeev Sharma announced the news on LinkedIn, revealing that this investment is part of a larger funding round. The company closed this round with a valuation of $151 million and plans to raise additional funds at a valuation of approximately $175 million. "With this new $4 million fundraise, which is part of a larger effort involving global investors, we will focus on developing new AI capabilities to tackle the challenges of autonomous general navigation," Sharma wrote. Previously, Sharma outlined the startup’s plans to enhance its R&D across all areas of autonomous driving, emphasizing unsupervised and inverse reinforcement learning to demonstrate their scalable Level-5 autonomous driving technology. Level 5 vehicles are fully autonomous, requiring no human intervention and lacking traditional controls like a steering wheel. In his post, Sharma emphasized the importance of addressing the most challenging AI problems to ensure safe and cost-effective autonomous driving. "Autonomous agents must learn to navigate the most complex, unpredictable, and adversarial traffic dynamics," he explained. The startup has created advanced AI and mathematical models to solve issues such as bidirectional traffic negotiation at speeds of 40 km/h on single-lane roads, managing multi-agent scenarios in dynamic environments, and navigating toll plazas and off-road areas at high speeds. Their technology also includes capabilities for LiDAR-less perception and unmanaged traffic-intersection negotiation both day and night. Swaayatt Robots previously raised about $3 million in 2021. Discussing their technology, Sharma said, "Our technology surpasses that of competitors in Europe, North America, and China. We have developed algorithmic frameworks that better handle environmental and traffic uncertainties while being more energy-efficient due to our computationally efficient algorithms." #SwaayattRobots #AutonomousDriving #AI #ArtificialIntelligence #Funding #FundingAlert #FundraiserAlert #AngelInvestors #TechInnovation #TechFunding #Level5Autonomy #SmartVehicles #Innovation #AutonomousVehicles #FutureOfMobility #MachineLearning #ReinforcementLearning #StartupGrowth #BhopalStartups #TransportationTech #TechBreakthroughs #AllBoutCorps

Swaayatt Robots, a Bhopal-based autonomous driving startup, has raised $4 million from US based angel investors. Sanjeev Sharma, Founder of Swaayatt Robots, took to LinkedIn to announce that the funding is part of a much larger round the startup is raising. The round was closed at a valuation of $151 million. Details ⬇️