This article concerns the opportunities for analysis of data about movement past spatial checkpoints. Checkpoint data are generated by an object’s movement past fixed sensors (‘checkpoints’) distributed throughout geographic space. Example sources of checkpoint data include road-toll gantries, social media check-ins, WiFi hotspots, access swipe cards, and public transport smart cards. Many existing movement analytics techniques, which frequently rely on precise coordinate locations, are ill-suited to the inherent and variable spatial granularity of checkpoint data. However, this spatial granularity also brings advantages in linking movement more closely to its environmental context, in particular the characteristics of the regions through which an object is moving. In this article, we propose a general model for representing checkpoint data of moving objects, based on two fundamentally different types of sensors: transaction and presence. Experiments using real movement data illustrate the diversity of queries that can be efficiently supported by our model. An example movement classification task, identifying vehicle type from checkpoint data recording vehicle movement, provides an illustration of the opportunities for more closely linking movement with its environmental context through the use of checkpoint data analysis.

Redundancy checking is an important task in AI subfields such as knowledge representation and constraint solving. This paper considers redundant topological constraints, defined in the region connection calculus RCC8. We say a constraint in a set C of RCC8 constraints is redundant if it is entailed by the rest of C. A prime subnetwork of C is a subset of C which contains no redundant constraints and has the same solution set as C. It is natural to ask how to compute such a prime subnetwork, and when it is unique. While this problem is in general intractable, we show that, if S is a subalgebra of RCC8 in which weak composition distributes over nonempty intersections, then C has a unique prime subnetwork, which can be obtained in cubic time by removing all redundant constraints simultaneously from C. As a by-product, we show that any path-consistent network over such a distributive subalgebra is minimal.

How does complex spatiotemporal behavior arise from, and from which, spatiotemporal knowledge? In an attempt to answer this question, we extend Valentino Braitenberg's thought experiment [3] by describing and implementing vehicles with explicit, and increasingly sophisticated, spatiotemporal knowledge. We then observe the corresponding spatiotemporal behavior that can result. These spatiotemporal vehicles are able to move about their environment. The paper shows how vehicles can be incrementally equipped with three fundamental spatial constructs: knowledge of places, of neighborhoods, and the ability to communicate with other nearby vehicles. In turn we demonstrate, using agent-based simulations, how the fundamental spatial concepts of fields, networks, objects, and reference frames can emerge from these basic constructs. Our approach contributes to ongoing efforts of identifying the core concepts of spatial information [11] and of understanding the relationships between interaction with space and spatial computation [6].

This paper examines efficient and decentralized monitoring of objects moving in a transportation network. Previous work in moving object monitoring has focused primarily on centralized information systems, like moving object databases and geographic information systems. In contrast, in this paper monitoring is in-network, requiring no centralized control and allowing for substantial spatial constraints to the movement of information. The transportation network is assumed to be augmented with fixed checkpoints that can detect passing mobile objects. This assumption is motivated by many practical applications, from traffic management in vehicle ad hoc networks to habitat monitoring by tracking animal movements. In this context, this paper proposes and evaluates a family of efficient decentralized algorithms for capturing, storing and querying the movements of objects. The algorithms differ in the restrictions they make on the communication and sensing constraints to the mobile nodes and the fixed checkpoints. The performance of the algorithms is evaluated and compared with respect to their scalability (in terms of communication and space complexity), and their latency (the time between when a movement event occurs, and when all interested nodes are updated with records about that event). The conclusions identify three key principles for efficient decentralized monitoring of objects moving past checkpoints: structuring computation around neighboring checkpoints; taking advantage of mobility diffusion and separating the generation and querying of movement information.