A Differential Game Theoretic Approach for Two-Agent Collision Avoidance with Travel Limitations

Goode, Brian J.; Roan, Michael J.

doi:10.1007/s10846-012-9657-x

Cited by 12 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, relying on the reward functions of MARL only is not sufficient to guarantee safety [7]. The collision-avoidance problem has been addressed based on different approaches, such as velocity obstacles [8], Voronoi partitions [9], differential games [10]. Nonetheless, the computational expense and the conservativeness of these approaches limit their applications in MARL.…”

Section: Introductionmentioning

confidence: 99%

Safe Multi-Agent Reinforcement Learning through Decentralized Multiple Control Barrier Functions

Cai,

Cao,

et al. 2021

Preprint

View full text Add to dashboard Cite

Multi-Agent Reinforcement Learning (MARL) algorithms show amazing performance in simulation in recent years, but placing MARL in real-world applications may suffer safety problems. MARL with centralized shields was proposed and verified in safety games recently. However, centralized shielding approaches can be infeasible in several real-world multi-agent applications that involve non-cooperative agents or communication delay. Thus, we propose to combine MARL with decentralized Control Barrier Function (CBF) shields based on available local information. We establish a safe MARL framework with decentralized multiple CBFs and develop Multi-Agent Deep Deterministic Policy Gradient (MADDPG) to Multi-Agent Deep Deterministic Policy Gradient with decentralized multiple Control Barrier Functions (MADDPG-CBF). Based on a collision-avoidance problem that includes not only cooperative agents but obstacles, we demonstrate the construction of multiple CBFs with safety guarantees in theory. Experiments are conducted and experiment results verify that the proposed safe MARL framework can guarantee the safety of agents included in MARL.

show abstract

Section: Introductionmentioning

confidence: 99%

Safe Multi-Agent Reinforcement Learning through Decentralized Multiple Control Barrier Functions

Cai,

Cao,

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…It is easy to give rise to automobile rear end collision accident, because of high speed, bad headway and the drivers' no timely reaction [1]. Statistical data show that more than eighty percent of accident is caused by driver's no timely reaction, and more than sixty five percent of vehicle collision belongs to rear end collision.…”

Section: Introductionmentioning

confidence: 99%

Automobile Active Collision and Early Warning System Based on DSP

Gao

Yuan

2013

AMM

View full text Add to dashboard Cite

To solve the problem of automobile rear end collision, an early warning system is designed, which is based on DSP chip control. Ultrasonic ranging is selected in the system. By providing pulse trigger signals, internal modules will send out eight periodic levels with forty kHz, and detect echoes at the same time. Once the echo signal is detected, the reverberation signal will be output. Then the reverberation signal can be changed into actual distance in real time by interrupt mode. The result stands for the distance between the car and the obstacles, then corresponding braking measures and early warning will be decided according to the minimum safety distance. When it is dangerous, the car status can be controlled according to the setting logic state.

show abstract