A survey on motion prediction and risk assessment for intelligent vehicles

By S. Lefèvre, Dizan Vasquez, Christian Laugier. 2014

1,122 citations recorded by the source corpus.

With the objective to improve road safety, the automotive industry is moving toward more “intelligent” vehicles. One of the major challenges is to detect dangerous situations and react accordingly in order to avoid or mitigate accidents. This requires predicting the likely evolution of the current traffic situation, and assessing how dangerous that future situation might be. This paper is a survey of existing methods for motion prediction and risk assessment for intelligent vehicles. The proposed classification is based on the semantics used to define motion and risk. We point out the tradeoff between model completeness and real-time constraints, and the fact that the choice of a risk assessment method is influenced by the selected motion model.

Subfields: Artificial Intelligence, Automotive Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, neutral): “[17] - - X - - - - - - - [18] - - - - - X X - - - [19] - - - - - X - - - - [20] X X X X - - - - - - [21] X - X - - - - - X - [22] X - - - - - - - - - [23] - - X X - X X - - X [24] - X - - X X X - - - [25] - - - - X X - - - - Ours X X X X X X - X X X complete the challenge attracted signiﬁcant attent”
An LSTM network for highway trajectory prediction (2017, background): “Many approaches to motion prediction have been proposed in the literature, and a survey can be found in [2].”
A Combined Model- and Learning-Based Framework for Interaction-Aware Maneuver Prediction (2016, methodology): “The survey of [4] classifies the existing methods of motion prediction in three different approaches: Physics-based, Maneuver-based and Interactionaware models.”
Human motion trajectory prediction: a survey (2019, methodology): “Our taxonomy offers a novel way to structure the growing body of literature, containing the categories proposed by Kruse et al. (2013), Lasota et al. (2017) and Lef ` evre et al. (2014) and extending them with a systematic categorization of contextual cues.”
Vehicle Trajectory Prediction by Integrating Physics- and Maneuver-Based Approaches Using Interactive Multiple Models (2018, background): “The physics-based models assume that some running parameters such as velocity and acceleration are constant in the near future [19].”
INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps (2019, background): “…these challenges, considerable amount of research efforts have been devoted to: i) predicting the future intention and motion of other road users [1]–[3], ii) modeling and analyzing driving behavior [4], [5], iii) clustering the motion and ﬁnding representation of the motion primitives [6], [7],…”
Real-time motion planning methods for autonomous on-road driving: State-of-the-art and future research directions (2015, methodology): “Interaction-aware models, as presented in Lefèvre et al. (2014) can take interaction between traffic participants into account, but pre-suppose perfect knowledge or communication between the vehicles.”
Risky Action Recognition in Lane Change Video Clips using Deep Spatiotemporal Networks with Segmentation Mask Transfer (2019, background): “This objective risk framework led to research focusing on vehicle tracking, where motion models predicted the future position of vehicles, allowing risk assessment [11], [12].”
A human-like game theory-based controller for automatic lane changing (2018, methodology): “These models can be classified into physics-based, maneuver-based and interaction-aware models (Lefèvre et al., 2014).”
Research Advances and Challenges of Autonomous and Connected Ground Vehicles (2019, methodology): “2) Motion Prediction: A detailed survey in [367] divides the existing motion prediction models into three categories: physics-based models, maneuver-based models, and interaction-aware models.”
Probabilistic vehicle trajectory prediction over occupancy grid map via recurrent neural network (2017, background): “So far, various approaches have been proposed to analyze the vehicles’ motion [1].”
Convolutional Social Pooling for Vehicle Trajectory Prediction (2018, methodology): “An extensive survey on vehicle motion prediction models has been presented by Lefevre et al. [16], where the the models are categorized into physics based, maneuver based and interaction aware models.”
A Survey on Trajectory-Prediction Methods for Autonomous Driving (2022, methodology): “Lefèvre et al. [4] present a survey on existing methods of motion prediction and risk assessment for AVs before 2014.”
Deep Learning-Based Vehicle Behavior Prediction for Autonomous Driving Applications: A Review (2020, methodology): “ed as the TV and its P(X TV jO EV ) is calculated, where: X TV = f(xT t ;y T );(xT +1 ;y T );:::;(xT +m ;y T )g (2) T 2f1;2;;Ngis the current TV. III. CLASSIFICATIONS OF PREVIOUS WORKS Lefevre et al. [12] classiﬁes vehicle behaviour prediction models to physics-based, manoeuvre-based, and interactio”
Interpretable End-to-End Urban Autonomous Driving With Latent Deep Reinforcement Learning (2020, methodology): “However, our method is able to obtain all those information within the end-to-end RL training process, without storing any HD maps or manually designing any localization algorithms.”
Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks (2018, background): “Comprehensive overview of the topic can be found in [17], [18], and in this section we review relevant work from the perspective of”
CoverNet: Multimodal Behavior Prediction Using Trajectory Sets (2019, background): “We focus on trajectory prediction approaches based on deep learning, and refer the reader to [26] for a survey of more classical approaches.”