Formalizing traffic rules for uncontrolled intersections

Karimi, Abolfazl; Duggirala, Parasara Sridhar

doi:10.1109/iccps48487.2020.00012

Cited by 27 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors of [20] focus on the translation of traffic rules from the California's DMV driver handbook from natural language to formal language (i.e. first order logic representations) and simulate their approach for exemplary four way and three way uncontrolled intersections using the CARLA urban driving simulator for autonomous vehicles [12].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Extending Urban Multi-Lane Spatial Logic to Formalise Road Junction Rules

Schwammberger

Alves

2021

Electron. Proc. Theor. Comput. Sci.

View full text Add to dashboard Cite

During the design of autonomous vehicles (AVs), several stages should include a verification process to guarantee that the AV is driving safely on the roads. One of these stages is to assure the AVs abide by the road traffic rules. To include road traffic rules in the design of an AV, a precise and unambiguous formalisation of these rules is needed. However, only recently this has been pointed out as an issue for the design of AVs and the few works on this only capture the temporal aspects of the rules, leaving behind the spatial aspects. Here, we extend the spatial traffic logic, Urban Multi-lane Spatial Logic, to formalise a subset of the UK road junction rules, where both temporal and spatial aspects of the rules are captured. Our approach has an abstraction level for urban road junctions that could easily promote the formalisation of the whole set of road junction rules and we exemplarily formalise three of the UK road junction rules. Once we have the whole set formalised, we will model, implement, and formally verify the behaviour of an AV against road traffic rules so that guidelines for the creation of a Digital Highway Code for AVs can be established. * This research was supported by the German Research Council (DFG) in the PIRE Projects SD-SSCPS and ISCE-ACPS under grants no. FR 2715/4-1, FR 2715/5-1. BackgroundWe give preliminary information about the approach on formalising traffic rules using temporal logic from [3] in Sect. 2.1 and in Sect. 2.2, we present details on the abstract model and logic Urban Multi-lane Spatial Logic UMLSL from [29]. The Road Junction RulesIn the UK Highway Code there are different sections which handle the road traffic rules for Overtaking, Roundabouts, Road Junctions, among others [11]. Here we are concerned with the

show abstract

Section: Related Workmentioning

confidence: 99%

“…They show that the behaviour of autonomous vehicles under their controller are more realistic compared to CARLA's default FIFO controller. However, neither of the aforementioned works formalise spatial aspects of traffic rules and only one of them [20] formalises road junction rules.…”

Section: Related Workmentioning

confidence: 99%

Extending Urban Multi-Lane Spatial Logic to Formalise Road Junction Rules

Schwammberger

Alves

2021

Electron. Proc. Theor. Comput. Sci.

View full text Add to dashboard Cite

show abstract

“…Additionally, use of CLINGO for executing ASP poses some limitations as discussed earlier (Gupta et al 2017). Karimi and Duggirala (Karimi and Duggirala 2020) have coded up rules from the California DMV handbook in ASP using CLINGO. Their goal is to verify the correctness of AV systems' behavior at intersections.…”

Section: Related Work and Conclusionmentioning

confidence: 99%

AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning

Kothawade¹,

Khandelwal²,

Basu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Driving an automobile involves the tasks of observing surroundings, then making a driving decision based on these observations (steer, brake, coast, etc.). In autonomous driving, all these tasks have to be automated. Autonomous driving technology thus far has relied primarily on machine learning techniques. We argue that appropriate technology should be used for the appropriate task. That is, while machine learning technology is good for observing and automatically understanding the surroundings of an automobile, driving decisions are better automated via commonsense reasoning rather than machine learning. In this paper, we discuss (i) how commonsense reasoning can be automated using answer set programming (ASP) and the goal-directed s(CASP) ASP system, and (ii) develop the AUTO-DISCERN 1 system using this technology for automating decision-making in driving. The goal of our research, described in this paper, is to develop an autonomous driving system that works by simulating the mind of a human driver. Since driving decisions are based on human-style reasoning, they are explainable, their ethics can be ensured, and they will always be correct, provided the system modeling and system inputs are correct.

show abstract

“…The structured program P, besides acting as the sequencer and trigger for the RL agents, also acts as a safety shield developed from safety specifications gleaned from well known driving rules. These kinds of safety specifications have been studied [13,34,35]. In symbolic model checking instead of enumerating reachable states of a state machine with embedded entry exit conditions one at a time, the state machine can be efficiently checked by taking a cross product with automaton constructed from the safety property (B üchi automaton).…”

Section: Safety Specificationsmentioning

confidence: 99%

Hierarchical Program-Triggered Reinforcement Learning Agents For Automated Driving

Gangopadhyay,

Soora,

Dasgupta

2021

Preprint

View full text Add to dashboard Cite

Recent advances in Reinforcement Learning (RL) combined with Deep Learning (DL) have demonstrated impressive performance in complex tasks, including autonomous driving [1]. The use of RL agents in autonomous driving leads to a smooth human-like driving experience, but the limited interpretability of Deep Reinforcement Learning (DRL) creates a verification and certification bottleneck. Instead of relying on RL agents to learn complex tasks, we propose HPRL -Hierarchical Program-triggered Reinforcement Learning, which uses a hierarchy consisting of a structured program along with multiple RL agents, each trained to perform a relatively simple task. The focus of verification shifts to the master program under simple guarantees from the RL agents, leading to a significantly more interpretable and verifiable implementation as compared to a complex RL agent. The evaluation of the framework is demonstrated on different driving tasks, and NHTSA precrash scenarios using CARLA, an open-source dynamic urban simulation environment.

show abstract

Formalizing traffic rules for uncontrolled intersections

Cited by 27 publications

References 9 publications

Extending Urban Multi-Lane Spatial Logic to Formalise Road Junction Rules

Extending Urban Multi-Lane Spatial Logic to Formalise Road Junction Rules

AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning

Hierarchical Program-Triggered Reinforcement Learning Agents For Automated Driving

Contact Info

Product

Resources

About