The motion of a tracked object often has long term underlying dependencies due to premeditated actions dictated by intent, such as destination. Revealing this intent, as early as possible, can enable advanced intelligent system functionalities for conflict/opportunity detection and automated decision making, for instance in surveillance and human computer interaction. This paper presents a novel Bayesian intent inference framework that utilises sequential Monte Carlo (SMC) methods to determine the destination of a tracked object exhibiting unknown jump behaviour. The latter can arise from the object undertaking fast maneuvers (e.g. for obstacle avoidance) and/or due to external uncontrollable environmental perturbations. Suitable intent-driven stochastic models and inference routines are introduced. The effectiveness of the proposed approach is demonstrated using synthetic and real data.

This letter presents an alternative, more consistent, construction for bridging distributions, which enables inferring the destination of a tracked object from the available partial sensory observations. Two algorithms are then introduced to sequentially estimate the probability of all possible endpoints within a generic Bayesian framework. They capture the influence of intended destination on the object's motion via suitably adapted stochastic models. Whilst the bridging approach has low training requirements, the proposed formulation can lead to more efficient predictors, e.g. around 65% less computations for certain models. Synthetic and real data is used to illustrate the effectiveness of the introduced algorithms.

In this paper, we present a Bayesian framework for manoeuvring object tracking and intent prediction using novel α-stable Lévy state-space models, expressed in continuous time as Lévy processes. In contrast to conventional (fully) Gaussian formulations, the proposed models are driven by heavy-tailed αstable noise and are thus much more able to capture extreme values/behaviours. This can better characterise sharp changes in the state, which may be induced by sudden and frequent manoeuvres such as swift turns or abrupt accelerations. In particular, they are represented in a conditionally Gaussian series form which ensures the tractability of the applied inference algorithms. A corresponding estimation strategy with the Rao-Blackwellised particle filter is then proposed and an efficient intent inference procedure is introduced. Here, the underlying intent, driving the target's long-term behaviour (e.g. reaching its final destination), is modelled as a latent variable. Real vessel data from maritime surveillance and human computer interactions (e.g. cursor data from motor-impaired interface users) are utilised to demonstrate the effectiveness of the proposed approach. It is shown to deliver noticeable improvements in the tracking and intent prediction performance (whenever relevant) compared with a more conventional Gaussian dynamic model.

In various scenarios, the motion of a tracked object, for example, a pointing apparatus, pedestrian, animal, vehicle, and others, is driven by achieving a premeditated goal such as reaching a destination. This is albeit the various possible trajectories to this endpoint. This paper presents a generic Bayesian framework that utilizes stochastic models that can capture the influence of intent (viz., destination) on the object behavior. It leads to simple algorithms to infer, as early as possible, the intended endpoint from noisy sensory observations, with relatively low computational and training data requirements. This framework is introduced in the context of the novel predictive touch technology for intelligent user interfaces and touchless interactions. It can determine, early in the interaction task or pointing gesture, the interface item the user intends to select on the display (e.g., touchscreen) and accordingly simplify as well as expedite the selection task. This is shown to significantly improve the usability of displays in vehicles, especially under the influence of perturbations due to road and driving conditions, and enable intuitive contact-free interactions. Data collected in instrumented vehicles are shown to demonstrate the effectiveness of the proposed intent prediction approach.

No abstract