Framework for Network-Constrained Tracking of Cyclists and Pedestrians

Abstract

The increase in perception capabilities of connected mobile sensor platforms (e.g., self-driving vehicles, drones, and robots) leads to an extensive surge of sensed features at various temporal and spatial scales. Beyond their traditional use for safe operation, available observations could enable to see how and where people move on sidewalks and cycle paths, to eventually obtain a complete microscopic and macroscopic picture of the traffic flows in a larger area. This paper proposes a new method for advanced traffic applications, tracking an unknown and varying number of moving targets (e.g., pedestrians or cyclists) constrained by a road network, using mobile (e.g., vehicles) spatially distributed sensor platforms. The key contribution in this paper is to introduce the concept of network bound targets into the multi-target tracking problem, and hence to derive a network-constrained multi-hypotheses tracker (NC-MHT) to fully utilize the available road information. This is done by introducing a target representation, comprising a traditional target tracking representation and a discrete component placing the target on a given segment in the network. A simulation study shows that the method performs well in comparison to the standard MHT filter in free space. Results particularly highlight network-constraint effects for more efficient target predictions over extended periods of time, and in the simplification of the measurement association process, as compared to not utilizing a network structure. This theoretical work also directs attention to latent privacy concerns for potential applications.

Prioritizing Cyclists at Signalized Intersections Using Observations from Connected Autonomous Vehicles

“…The way iTLCs process the input from sensor platforms to grant priority to individual cyclists, or a platoon of cyclists, is comparable to the way data from static cameras or radar are handled. An example where bicycle priority is applied using smart cameras detecting bicycle platoons of three or more cyclists can be found in the municipality of Hengelo (NL) ( 19 ), whereas bicycle priority using cyclist observations from static radar was applied in the municipality of Delft (NL) ( 12 ).…”

“…We used the network-constrained multiple hypothesis tracking (NC-MHT) approach proposed in research by Vial et al ( 19 ) to update the information about a target’s position along a defined road network from a temporal sequence of observations.…”

“…A complete solution to the MTT problem is given by a global tracking hypothesis, which represents the inferred number of targets and their respective states along the road network. We refer the interested reader to the original work ( 19 ) for more details on, for example, data association logic, track likelihood computations, or handling of the road constraints, as we focus on the target state description and estimation models for the rest of this section.…”

“…If no observation is available, the update step is simply skipped, and if several are available, the step is repeated. We refer the interested reader to Vial et al’s study ( 19 ) for more details on network-constrained effects and assumptions in the context of the NC-MHT framework.…”

Improved generalization of cyclist detection on security cameras with the OpenImages Cyclists dataset

“…Other approaches aimed at pedestrian and cyclist safety on public roads are presented in [Vial et al, 2023], with the use of mobile sensors for traffic tracking applications, and in [Abadi et al, 2022], [Fang and López, 2020] and [Ahmed et al, 2019] who propose prediction of cyclist behavior by autonomous vehicles or driver assistance systems, using deep neural networks, which investigate movement tracking and pose estimation. In [Pool et al, 2019] the authors propose a neural network that identifies contextual information for cyclist path prediction by autonomous vehicles…”