MADDPG: Multi-agent Deep Deterministic Policy Gradient Algorithm for Formation Elliptical Encirclement and Collision Avoidance

Xu, Leixin; Chen, Weibin; Liu, Xiang; Chen, Yangyang

doi:10.1007/978-981-19-3998-3_24

Cited by 3 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The policy of an agent (UAV) is trained in a try-and-fail manner through repeated interactions with a simulation environment. Multi-Agent Reinforcement Learning (MARL) trains cooperative policies for agents with a Centralized Training with Decentralized Execution (CTDE) scheme [20], [21], [22], [23]. In detail, agents are trained in a centralized manner considering others agents' policies.…”

Section: Reinforcement Learning (Rl) Studies the Problem What Tomentioning

confidence: 99%

“…It will help if the collisions occurring in future steps can be foreseen. Recently, Multi-Agent Reinforcement Learning (MARL) has been used to train policies for cooperative collision avoidance [20], [21], [22], [23]. Policies are trained to plan trajectories at each step considering long-term consequences to address the shortsight limitation of shifting horizon planning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

Huang¹,

Zhang²,

Huang

2021

ICAPS

View full text Add to dashboard Cite

Energy efficiency is of critical importance to trajectory planning for UAV swarms in obstacle avoidance. In this paper, we present E2Coop, a new scheme designed to avoid collisions for UAV swarms by tightly coupling Artificial Potential Field (APF) with Particle Swarm Planning (PSO) based trajectory planning. In E2Coop, swarm members perform trajectory planning cooperatively to avoid collisions in an energy-efficient manner. E2Coop exploits the advantages of the active contour model in image processing for trajectory planning. Each swarm member plans its trajectories on the contours of the environment field to save energy and avoid collisions to obstacles. Swarm members that fall within the safeguard distance of each other plan their trajectories on different contours to avoid collisions with each other. Simulation results demonstrate that E2Coop can save energy up to 51% compared with two state-of-the-art schemes.

show abstract

Section: Reinforcement Learning (Rl) Studies the Problem What Tomentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

Huang¹,

Zhang²,

Huang

2021

ICAPS

View full text Add to dashboard Cite

show abstract

Large-scale UAVs autonomous rendezvous Using partially observed mean-field reinforcement learning

Zhang,

Ding,

et al. 2024

2024 IEEE 18th International Conference on Control &Amp;amp; Automation (ICCA)

View full text Add to dashboard Cite

E ² CoPre: Energy Efficient and Cooperative Collision Avoidance for UAV Swarms With Trajectory Prediction

Huang,

Zhang,

Huang

2024

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

MADDPG: Multi-agent Deep Deterministic Policy Gradient Algorithm for Formation Elliptical Encirclement and Collision Avoidance

Cited by 3 publications

References 19 publications

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

Large-scale UAVs autonomous rendezvous Using partially observed mean-field reinforcement learning

E ² CoPre: Energy Efficient and Cooperative Collision Avoidance for UAV Swarms With Trajectory Prediction

Contact Info

Product

Resources

About

MADDPG: Multi-agent Deep Deterministic Policy Gradient Algorithm for Formation Elliptical Encirclement and Collision Avoidance

Cited by 3 publications

References 19 publications

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

E2Coop: Energy Efficient and Cooperative Obstacle Detection and Avoidance for UAV Swarms

Large-scale UAVs autonomous rendezvous Using partially observed mean-field reinforcement learning

E 2 CoPre: Energy Efficient and Cooperative Collision Avoidance for UAV Swarms With Trajectory Prediction

Contact Info

Product

Resources

About

E ² CoPre: Energy Efficient and Cooperative Collision Avoidance for UAV Swarms With Trajectory Prediction