Autonomy 2.0: Why is self-driving always 5 years away?

Jain, Anurag; Pero, Luca Del; Grimmett, Hugo; Ondrúška, Peter

doi:10.48550/arxiv.2107.08142

Cited by 7 publications

(6 citation statements)

References 24 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Imitation Learning In Ashesh et al [5], the basis of the work relies on imitation learning -using human demonstrations to train a model. This method is promising, but it requires a significant number of training samples from real-world data.…”

Section: Related Workmentioning

confidence: 99%

“…dist((x t , y t ), (midpoint(p i , p i+1 )) (5) where dist and midpoint are the formulas to compute the Euclidean distance between two points and finding the midpoint between two points, respectively. In our implementation, we compute this by looping through each consecutive pair of polyline points while calculating the distance of the vehicle from the segment's midpoint and returning the smallest of distances between the vehicle position.…”

Section: Loss Functionmentioning

confidence: 99%

See 1 more Smart Citation

Designing a Recurrent Neural Network to Learn a Motion Planner for High-Dimensional Inputs

Chiu¹

2022

Preprint

View full text Add to dashboard Cite

The use of machine learning in the self-driving industry has boosted a number of recent advancements. In particular, the usage of large deep learning models in the perception and prediction stack have proved quite successful, but there still lacks significant literature on the use of machine learning in the planning stack. The current state of the art in the planning stack often relies on fast constrained optimization or rule-based approaches. Both of these techniques fail to address a significant number of fundamental problems that would allow the vehicle to operate more similarly to that of human drivers. In this paper, we attempt to design a basic deep learning system to approach this problem. Furthermore, the main underlying goal of this paper is to demonstrate the potential uses of machine learning in the planning stack for autonomous vehicles (AV) and provide a baseline work for ongoing and future research.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Loss Functionmentioning

confidence: 99%

Designing a Recurrent Neural Network to Learn a Motion Planner for High-Dimensional Inputs

Chiu¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Completely replacing human drivers with artificial drivers is indeed desirable, but still a challenging task. Production-level deployment of full self-driving vehicles remains a distant future (Jain et al, 2021 ). On the one hand, state-of-the-art driving agents surpass humans in computation, responsiveness, and multitasking.…”

Section: Vehicle Intelligence With Single Cognitionmentioning

confidence: 99%

Distributed cognition for collaboration between human drivers and self-driving cars

Plebe

Papini

Cherubini

et al. 2022

Front. Artif. Intell.

View full text Add to dashboard Cite

This paper focuses on the collaboration between human drivers and intelligent vehicles. We propose a collaboration mechanism grounded on the concept of distributed cognition. With distributed cognition, intelligence does not lie just in the single entity but also in the interaction with the other cognitive components in a system. We apply this idea to vehicle intelligence, proposing a system distributed into two cognitive entities—the human and the autonomous agent—that together contribute to drive the vehicle. This account of vehicle intelligence differs from the mainstream research effort on highly autonomous cars. The proposed mechanism follows one of the paradigm derived from distributed cognition, the rider-horse metaphor: just like the rider communicates their intention to the horse through the reins, the human influences the agent using the pedals and the steering wheel. We use a driving simulator to demonstrate the collaboration in action, showing how the human can communicate and interact with the agent in various ways with safe outcomes.

show abstract

“…While the data-driven approach to autonomous driving is viewed as the most promising path to full autonomy by some authors [18,16], significant problems remain to be solved, most prominently in generalization, in the explainability of decisions, and in providing safety guarantees [9,24,19].…”

Section: Introductionmentioning

confidence: 99%

Controlling Steering with Energy-Based Models

Mykyta¹,

Tampuu²,

Matiisen³

2023

Preprint

View full text Add to dashboard Cite

So-called implicit behavioral cloning with energy-based models has shown promising results in robotic manipulation tasks. We tested if the method's advantages carry on to controlling the steering of a real self-driving car with an end-to-end driving model. We performed an extensive comparison of the implicit behavioral cloning approach with explicit baseline approaches, all sharing the same neural network backbone architecture. Baseline explicit models were trained with regression (MAE) loss, classification loss (softmax and cross-entropy on a discretization), or as mixture density networks (MDN). While models using the energy-based formulation performed comparably to baseline approaches in terms of safety driver interventions, they had a higher whiteness measure, indicating higher jerk. To alleviate this, we show two methods that can be used to improve the smoothness of steering. We confirmed that energy-based models handle multimodalities slightly better than simple regression, but this did not translate to significantly better driving ability. We argue that the steering-only road-following task has too few multimodalities to benefit from energy-based models. This shows that applying implicit behavioral cloning to real-world tasks can be challenging, and further investigation is needed to bring out the theoretical advantages of energy-based models.

show abstract

Autonomy 2.0: Why is self-driving always 5 years away?

Cited by 7 publications

References 24 publications

Designing a Recurrent Neural Network to Learn a Motion Planner for High-Dimensional Inputs

Designing a Recurrent Neural Network to Learn a Motion Planner for High-Dimensional Inputs

Distributed cognition for collaboration between human drivers and self-driving cars

Controlling Steering with Energy-Based Models

Contact Info

Product

Resources

About