Safety is crucial for robotic missions within an uncertain environment. Common safety requirements such as collision avoidance are only state-dependent, which can be restrictive for complex missions. In this work, we address a more general formulation as safe-return constraints, which require the existence of a return-policy to drive the system back to a set of safe states with high probability. The robot motion is modeled as a Markov Decision Process (MDP) with probabilistic labels, which can be highly non-ergodic. The robotic task is specified as Linear Temporal Logic (LTL) formulas over these labels, such as surveillance and transportation. We first provide theoretical guarantees on the re-formulation of such safe-return constraints, and a baseline solution based on computing two complete product automata. Furthermore, to tackle the computational complexity, we propose a hierarchical planning algorithm that combines the feature-based symbolic and temporal abstraction with constrained optimization. It synthesizes simultaneously two dependent motion policies: the outbound policy minimizes the overall cost of satisfying the task with a high probability, while the return policy ensures the safe-return constraints. The problem formulation is versatile regarding the robot model, task specifications and safety constraints. The proposed hierarchical algorithm is more efficient and can solve much larger problems than the baseline solution, with only a slight loss of optimality. Numerical validations include simulations and hardware experiments of a search-and-rescue mission and a planetary exploration mission over various system sizes.

Between discarded packaging, shipping fees, inventory shortages and damaged merchandise, returns cost retailers a fortune. In 2017, retail returns cost roughly $350 billion, and they were projected to rise to $550 million in 2020. While some returns are inevitable (with reasons ranging from defective products to late deliveries), others are entirely avoidable. In fact, 46% of shoppers surveyed in Narvar's 2019 The State of Online Returns report said their No. 1 reason for returning products was incorrect size, fit or color. Only 3% intentionally bought multiple items knowing they'd return some.

The emergence of AI technology is the result of evolved human thinking. And, in some ways, it's become one of the biggest technology game changers of this generation. However, without being very cautious and aware of the dangers--and significance--of combining human interaction with artificial intelligence, we could head down the wrong path. Personally, a substantial barrier I've faced in my life is properly managing my time. At first, I was obsessed with learning the ins and outs of Google Calendar.

When attending trade shows and conferences supporting Conversica, a question I am frequently asked is whether Conversica's Intelligent Virtual Assistant for customer engagement is "a chatbot." And while I can understand the source of the question, I emphatically stress that Conversica is not a chatbot. A more cynical reader might assume that this differentiation is little more than branding. But I can assure you there are very real differences between what the Conversica Sales AI Assistant offers and what chatbot providers deliver. Each technology has its place and purpose, but neither is synonymous with the other.

In a post on its corporate blog on Monday, eBay explained how it uses machine learning technology to drive innovative new approaches to search experiences. But while that may sound good to the rest of the world, sellers may be wary of marketplaces that tinker with search, which is key to getting exposure to shoppers. Even before eBay started touting its skills in artificial intelligence and machine learning, sellers were concerned over eBay's use of an algorithm to influence which listings would appear higher on the search results page, something known as Best Match - the default sort order of listings. And that's exactly where eBay is concentrating its testing. In Monday's post, eBay wrote, "The largest scale application of machine learning technology at eBay is currently Best Match, the algorithm used to optimize relevance for buyers during their shopping experiences. It is a powerful tool for surfacing deals."