Abstract
Autonomous vehicle (AV) industry has evolved rapidly during the past decade. Research and development in each sub-module (perception, state estimation, motion planning etc.) of AVs has seen a boost, both on the hardware (variety of new sensors) and the software sides (state-of-the-art algorithms). With recent advancements in achieving real-time performance using onboard computational hardware on an ego vehicle, one of the major challenges that AV industry faces today is modelling behaviour and predicting future intentions of road users. To make a self-driving car reason and execute the safest motion plan, it should be able to understand its interactions with other road users. Modelling such behaviour is not trivial and involves various factors e.g. demographics, number of traffic participants, environmental conditions, traffic rules, contextual cues etc. This comprehensive review summarizes the related literature. Specifically, we identify and classify motion prediction literature for two road user classes i.e. pedestrians and vehicles. The taxonomy proposed in this review gives a unified generic overview of the pedestrian and vehicle motion prediction literature and is built on three dimensions i.e. motion modelling approach, model output type, and situational awareness from the perspective of an AV.
Highlights
S AFETY Safety is a crucial aspect for an autonomous vehicle (AV)
The taxonomy proposed in this review gives a unified generic overview of the pedestrian and vehicle motion prediction literature and is built on three dimensions i.e. motion modelling approach, model output type, and situational awareness from the perspective of an Autonomous vehicle (AV)
This means that for an AV to coexist with other road users, it should follow the traffic rules and regulations and be socially aware i.e. it should understand the interactions of road users to ensure the flow of traffic [1]
Summary
S AFETY Safety is a crucial aspect for an autonomous vehicle (AV). Other road users that an AV needs to interact with, come in many forms. We are interested in categorizing state-of-the-art literature into a novel taxonomy that incorporates motion prediction methods of the two road user classes i.e. pedestrians and vehicles For these road user types, this work classifies the literature on the basis of modelling approach (physics, learning-based), output type (trajectories, intentions etc.) and situational awareness (interactions with scene objects). CNNs are commonly used for extracting features from images; later these features can be passed to a fully connected network to obtain some useful output which, in the case of behaviour prediction, can be a trajectory, or an occupancy or transition map An example of such a network employed for the case of pedestrians is [36] where the authors claim that CNNs can do better in capturing long-term temporal dependencies compared to LSTMs. Authors in [37] used human and machine annotated images to forecast pedestrian movement using a model built on resNet [38]. Object detection, tracking and motion forecasting are performed in an end-to-end fashion
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
