2015
DOI: 10.1007/978-3-319-16595-0_10
|View full text |Cite
|
Sign up to set email alerts
|

Real-Time Predictive Modeling and Robust Avoidance of Pedestrians with Uncertain, Changing Intentions

Abstract: Abstract. To plan safe trajectories in urban environments, autonomous vehicles must be able to quickly assess the future intentions of dynamic agents. Pedestrians are particularly challenging to model, as their motion patterns are often uncertain and/or unknown a priori. This paper presents a novel changepoint detection and clustering algorithm that, when coupled with offline unsupervised learning of a Gaussian process mixture model (DPGP), enables quick detection of changes in intent and online learning of mo… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
38
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
7
1

Relationship

2
6

Authors

Journals

citations
Cited by 43 publications
(38 citation statements)
references
References 24 publications
0
38
0
Order By: Relevance
“…The pedestrian position is measured using a lidar sensor placed in front of the intersection. For more details of the experimental setup, please refer to [33] or [34,Chs. 5 and 6].…”
Section: Pedestrian Behavior Classificationmentioning
confidence: 99%
“…The pedestrian position is measured using a lidar sensor placed in front of the intersection. For more details of the experimental setup, please refer to [33] or [34,Chs. 5 and 6].…”
Section: Pedestrian Behavior Classificationmentioning
confidence: 99%
“…Step 1) can be computed in a probabilistic sense using (5). More sophisticated methods have also been developed to detect changes in an agent's intention [31].…”
Section: Real-time Vision-based Path Planningmentioning
confidence: 99%
“…Then, we can make predictions by finding the set of possible subsequent dictionary atoms {d j |T kj > 0}. We compare our algorithm against using Gibbs sampling for the Dirichlet Process Gaussian Process (DPGP) model [14], which clusters training trajectories with a DP prior and models each cluster using a GP. In contrast, we model each transition (concatenation of two dictionary atoms {d k , d j |T kj > 0}) as a GP.…”
Section: B Making Predictions At An Intersectionmentioning
confidence: 99%
“…Learning local motion patterns is contrasted with assigning global cluster labels to each trajectory as in clusteringbased methods [7], [14]. We provide a detailed comparison between these two views in section IV-B.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation