Modeling intent and destination prediction within a Bayesian framework: Predictive touch as a usecase

Abstract

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.

Infra‐Red Intelligent Surface for Near‐Field Touchless Displays

“…). Cambridge University conducted research on predictive touch with Jaguar Land Rover, 4,5 and with other major car companies implementing the same, predictive touch is becoming a hot topic. To our knowledge, only Embodme has a fully integrated in‐screen solution, with a clear and readable interaction area capable of solving ergonomic and accessibility issues.…”

Detection of Malicious Intent in Non-cooperative Drone Surveillance

“…Existing model-based intent inference techniques in object tracking are largely focused on: 1) determining the target final destination out of a finite set of nominal endpoints at known locations, e.g. [9]- [13], see [14], [15] for an overview; 2) destination-aware tracking based on reciprocal processes [16] for pre-defined endpoints; 3) long-term trajectory forecasting using motion models learnt in advance, e.g. [17], [18].…”

“…a continuous intent space), e.g. [9]- [15], or require an off-line training phase for the learning of intent parameters [17], [18].…”

“…The latter can dynamically change over time here and the destination sought by the object/UAS can be an extended spatial region of any shape, located anywhere in the surveyed area. Thus, it does not require the prior knowledge of a set of possible endpoints as in [9]- [15],…”

“…Compared to examining specific spatial points or locations, e.g. in [9], [15], calculating the probability that geographical area(s) or any point/region within it/them in this paper is a more generic and fitting formulation for the tackled problem of early prediction of a threatening or anomalous drone behaviour.…”