A key open challenge in off-road autonomy is that the traversability of terrain often depends on the vehicle's state. In particular, some obstacles are only traversable from some orientations. However, learning this interaction by encoding the angle of approach as a model input demands a large and diverse training dataset and is computationally inefficient during planning due to repeated model inference. To address these challenges, we present SPARTA, a method for estimating approach angle conditioned traversability from point clouds. Specifically, we impose geometric structure into our network by outputting a smooth analytical function over the 1-Sphere that predicts risk distribution for any angle of approach with minimal overhead and can be reused for subsequent queries. The function is composed of Fourier basis functions, which has important advantages for generalization due to their periodic nature and smoothness. We demonstrate SPARTA both in a high-fidelity simulation platform, where our model achieves a 91\% success rate crossing a 40m boulder field (compared to 73\% for the baseline), and on hardware, illustrating the generalization ability of the model to real-world settings. Our code will be available at https://github.com/neu-autonomy/SPARTA.

The app leverages massive training data from millions of images to see vehicle damage in the same manner as a human assessor. AI-based visual analysis has applications in a broad range of fields. It can be used on an assembly line to inspect for product defects, detect the difference between a human and an animal, and more. Tractable, a visual AI company, has found an interesting use that assesses a vehicle condition. The recently launched artificial intelligence solution is designed to assess the condition of the external vehicle body quickly and effectively.

Whether you envisage self-driving vehicles to be five or 25 years away, given the substantial investments and technological progress being made, it is hard to argue against the industry's narrative of inevitability. Not to be outdone in this digital age, insurance claims is forging its own digital transformation path, leveraging advancements in technology and analytics to inject automation into the claim process. So how far away are we from delivering claim processing that is automated end-to-end? Will this happen before driverless Ubers are roaming the streets?

Smart IT systems are now calculating claims costs and attributing fault for accidents without any human involvement, speeding up the resolution of claims. Technology is set to transform motor insurance in the next five to 10 years, revolutionising both the claims process and repair. Artificial intelligence (AI) is enabling insurers to evaluate vehicle damage at the scene of a collision, without the need for a claims handler or loss adjustor. By analysing millions of photos of vehicle damage and cross-referencing them with actual repairs, programmers have been able to create algorithms that can assess the scale of the damage and create a full estimate including recommended repair, paint, parts costs and labour hours. The system can determine, for example, whether body panels can be repaired or need replacing, and in worse case scenarios it ensures that no total losses are sent to bodyshops.

Machine learning could be used to underwrite faster, reduce fraud and assess vehicle damages more accurately, Lawrence Wong, Munich Re Canada's director of application development, said Tuesday. Technology Conference in Toronto, Wong said that artificial intelligence (AI) is here to stay and will augment the way people do the business of insurance, whether insurers leverage it or not. "The power of machine learning is that you don't have to explicitly program it," Wong said during the session Case Studies in AI. "As you show [a self-learning machine] more and more pictures of bumper damage, it will, with experience, become more accurate at defining what's vehicle damage versus what's quirky vehicle design from the car manufacturer. "Notice I said the word experience," Wong added. "That's exactly how a human loss adjuster would learn about vehicle damage.