Transferable Pedestrian Motion Prediction Models at Intersections

نویسندگان

  • Macheng Shen
  • Golnaz Habibi
  • Jonathan P. How
چکیده

One desirable capability of autonomous cars is to accurately predict the pedestrian motion near intersections for safe and efficient trajectory planning. We are interested in developing transfer learning algorithms that can be trained on the pedestrian trajectories collected at one intersection and yet still provide accurate predictions of the trajectories at another, previously unseen intersection. We first discussed the feature selection for transferable pedestrian motion models in general. Following this discussion, we developed one transferable pedestrian motion prediction algorithm based on Inverse Reinforcement Learning (IRL) that infers pedestrian intentions and predicts future trajectories based on observed trajectory. We evaluated our algorithm on a dataset collected at two intersections, trained at one intersection and tested at the other intersection. We used the accuracy of augmented semi-nonnegative sparse coding (ASNSC), trained and tested at the same intersection as a baseline. The result shows that the proposed algorithm improves the baseline accuracy by 40% in the non-transfer task, and 16% in the transfer task.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Priority Order for Improvement of Intersections using Pedestrian Crash Prediction Model

One of the most important needs of pedestrians is safety at crossing points, especially at intersections. Intersections are important parts of the urban road network because any disruption in them reduces the capacity of the entire network. The main objective of this research is to propose an appropriate method for prioritizing urban intersections with considering the important factors affectin...

متن کامل

CASNSC: A context-based approach for accurate pedestrian motion prediction at intersections

Intention recognition of pedestrians is crucial to safe and reliable working of 1 autonomous vehicles, when serving as, for instance, indoor service robots or 2 self-driving cars in busy urban scenes. Previously, Chen et al. [2016] combined 3 Markovian-based and clustering-based approaches to learn motion primitives and 4 subsequently predict pedestrian trajectories by modeling the transition b...

متن کامل

CASNSC: A context-based approach for accurate pedestrian motion prediction at intersections

Intention recognition of pedestrians is crucial to safe and reliable working of autonomous vehicles, when serving as, for instance, indoor service robots or self-driving cars in busy urban scenes. Previously, Chen et al. [2016] combined Markovian-based and clustering-based approaches to learn motion primitives and subsequently predict pedestrian trajectories by modeling the transition between l...

متن کامل

Estimating Pedestrian Volumes and Crashes at Urban Signalized Intersections

Crash prediction models are used to estimate the number of crashes using a set of explanatory variables. The highway safety community has used modeling techniques to predict vehicle-to-vehicle crashes for decades. Specifically, generalized linear models (GLMs) are commonly used because they can model non-linear count data such as motor vehicle crashes. Regression models such as the Poisson, Zer...

متن کامل

Contextual Combination of Appearance and Motion for Intersection Videos with Vehicles and Pedestrians

Object detection and classification is challenging problem for vision-based intersection monitoring since traditional motion-based techniques work poorly when pedestrians or vehicles stop due to traffic signals. In this work, we present a method for vehicle and pedestrian recognition at intersections that benefits from both motion and appearance cues in video surveillance. Vehicle and pedestria...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2018