English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Finding feasible and collision-free paths for multiple nonlinear agents ischallenging in the decentralized scenarios due to limited available informationof other agents and complex dynamics constraints. In this paper, we propose afast multi-agent collision avoidance algorithm for general nonlinear agentswith continuous action space, where each agent observes only positions andvelocities of nearby agents. To reduce online computation, we first decomposethe multi-agent scenario and solve a two agents collision avoidance problemusing reinforcement learning (RL). When extending the trained policy to amulti-agent problem, safety is ensured by introducing the optimal reciprocalcollision avoidance (ORCA) as linear constraints and the overall collisionavoidance action could be found through simple convex optimization. Mostexisting RL-based multi-agent collision avoidance algorithms rely on the directcontrol of agent velocities. In sharp contrasts, our approach is applicable togeneral nonlinear agents. Realistic simulations based on nonlinear bicycleagent models are performed with various challenging scenarios, indicating acompetitive performance of the proposed method in avoiding collisions,congestion and deadlock with smooth trajectories.

Reciprocal Collision Avoidance for General Nonlinear Agents using Reinforcement Learning