Worst-case Analysis of the Time-To-React Using Reachable Sets

Sontges, Sebastian; Koschi, Markus; Althoff, Matthias

doi:10.1109/ivs.2018.8500709

Cited by 27 publications

(18 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since collision prediction is key to the safety of AVs, a wide range of solutions have been proposed by academia and industry. As mentioned earlier, current consumer vehicles use statistics-based SBTMs for collision prediction but can perform poorly in complex situations [10], [11] or react too late to avoid collisions [12], [13]. Expanding on these approaches, companies like Mobileye and Nvidia have proposed more comprehensive mathematical models for ensuring AV safety, namely Responsibility-Sensitive Safety (RSS) [22] and Nvidia Safety Force Field [23], respectively.…”

Section: A Early Collision Predictionmentioning

confidence: 99%

“…As a result, these methods are less capable of generalizing and can perform poorly in complex road scenarios. Moreover, to reduce false positives, these systems are designed to respond at the last possible moment [12]. Under such circumstances, the AV control system can fail to take timely corrective actions [13] if the system fails to predict a collision or estimates the TTC inaccurately.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatio-Temporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

Malawade¹,

Yu²,

Hsu³

et al. 2021

Preprint

View full text Add to dashboard Cite

In autonomous vehicles (AV), early warning systems rely on collision prediction to ensure occupant safety. However, state-of-the-art methods using deep convolutional networks either fail at modeling collisions or are too expensive/slow, making them less suitable for deployment on AV edge hardware. To address these limitations, we propose SG2VEC, a spatio-temporal scenegraph embedding methodology that uses Graph Neural Network (GNN) and Long Short-Term Memory (LSTM) layers to predict future collisions via visual scene perception. We demonstrate that SG2VEC predicts collisions 8.11% more accurately and 39.07% earlier than the state-of-the-art method on synthesized datasets, and 29.47% more accurately on a challenging real-world collision dataset. We also show that SG2VEC is better than the state-ofthe-art at transferring knowledge from synthetic datasets to realworld driving datasets. Finally, we demonstrate that SG2VEC performs inference 9.3x faster with an 88.0% smaller model, 32.4% less power, and 92.8% less energy than the state-of-the-art method on the industry-standard Nvidia DRIVE PX 2 platform, making it more suitable for implementation on the edge.

show abstract

Section: A Early Collision Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Spatio-Temporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

Malawade¹,

Yu²,

Hsu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Let T i (t) be the instantaneous time-to-collision between the ego vehicle and the i-th environment vehicle at time step t. The value T i (t) can be defined in multiple ways (see e.g. Sontges et al [79]). Norden et al [63] define it as the amount of time that would elapse before the two vehicles' bounding boxes intersect assuming that they travel at constant fixed velocities from the snapshot at time t. Time-to-collision captures directly whether or not the ego-vehicle was involved in a crash.…”

Section: B8 Proof Of Lemmamentioning

confidence: 99%

FormulaZero: Distributionally Robust Online Adaptation via Offline Population Synthesis

Sinha¹,

O’Kelly²,

Zheng³

et al. 2020

Preprint

View full text Add to dashboard Cite

Balancing performance and safety is crucial to deploying autonomous vehicles in multiagent environments. In particular, autonomous racing is a domain that penalizes safe but conservative policies, highlighting the need for robust, adaptive strategies. Current approaches either make simplifying assumptions about other agents or lack robust mechanisms for online adaptation. This work makes algorithmic contributions to both challenges. First, to generate a realistic, diverse set of opponents, we develop a novel method for self-play based on replicaexchange Markov chain Monte Carlo. Second, we propose a distributionally robust bandit optimization procedure that adaptively adjusts risk aversion relative to uncertainty in beliefs about opponents behaviors. We rigorously quantify the tradeoffs in performance and robustness when approximating these computations in real-time motion-planning, and we demonstrate our methods experimentally on autonomous vehicles that achieve scaled speeds comparable to Formula One racecars.

show abstract

“…The value T i (t) can be defined in multiple ways (see e.g. Sontges et al [49]). In this study, we define it as the amount of time that would elapse before the two vehicles' bounding boxes intersect assuming that they travel at constant fixed velocities from the snapshot at time t.…”

Section: A Simulation Parametersmentioning

confidence: 99%

Efficient Black-box Assessment of Autonomous Vehicle Safety

Norden,

O'Kelly,

Sinha

2019

Preprint

View full text Add to dashboard Cite

While autonomous vehicle (AV) technology has shown substantial progress, we still lack tools for rigorous and scalable testing. Real-world testing, the de-facto evaluation method, is dangerous to the public. Moreover, due to the rare nature of failures, billions of miles of driving are needed to statistically validate performance claims. Thus, the industry has largely turned to simulation to evaluate AV systems. However, having a simulation stack alone is not a solution. A simulation testing framework needs to prioritize which scenarios to run, learn how the chosen scenarios provide coverage of failure modes, and rank failure scenarios in order of importance. We implement a simulation testing framework that evaluates an entire modern AV system as a black box. This framework estimates the probability of accidents under a base distribution governing standard traffic behavior. In order to accelerate rare-event probability evaluation, we efficiently learn to identify and rank failure scenarios via adaptive importance-sampling methods. Using this framework, we conduct the first independent evaluation of a full-stack commercial AV system, Comma AI's OpenPilot.

show abstract

Worst-case Analysis of the Time-To-React Using Reachable Sets

Cited by 27 publications

References 20 publications

Spatio-Temporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

Spatio-Temporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

FormulaZero: Distributionally Robust Online Adaptation via Offline Population Synthesis

Efficient Black-box Assessment of Autonomous Vehicle Safety

Contact Info

Product

Resources

About