{"id":27040022,"url":"https://github.com/cdvel/hlea4tc","last_synced_at":"2026-03-11T19:33:48.684Z","repository":{"id":31409164,"uuid":"34972471","full_name":"cdvel/HLEA4TC","owner":"cdvel","description":"Multiagent communication artifact, with JADE and JNI (Hierarchical Reinforcement Learner)","archived":false,"fork":false,"pushed_at":"2025-05-28T05:42:09.000Z","size":93,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-08-19T03:35:38.364Z","etag":null,"topics":["research"],"latest_commit_sha":null,"homepage":"","language":"Java","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/cdvel.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2015-05-03T03:32:02.000Z","updated_at":"2025-05-28T05:42:23.000Z","dependencies_parsed_at":"2025-06-15T08:39:07.585Z","dependency_job_id":null,"html_url":"https://github.com/cdvel/HLEA4TC","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/cdvel/HLEA4TC","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cdvel%2FHLEA4TC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cdvel%2FHLEA4TC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cdvel%2FHLEA4TC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cdvel%2FHLEA4TC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cdvel","download_url":"https://codeload.github.com/cdvel/HLEA4TC/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cdvel%2FHLEA4TC/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30395669,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-11T18:46:22.935Z","status":"ssl_error","status_checked_at":"2026-03-11T18:46:17.045Z","response_time":84,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["research"],"created_at":"2025-04-05T03:27:35.653Z","updated_at":"2026-03-11T19:33:48.667Z","avatar_url":"https://github.com/cdvel.png","language":"Java","funding_links":[],"categories":[],"sub_categories":[],"readme":"# HLEA4TC - Hierarchical Learning Engine for Adaptive Traffic Control\n\nA JADE-based multi-agent framework designed for distributed traffic control optimization with support for hierarchical agent organization and external traffic simulation integration.\n\n## Overview\n\nHLEA4TC provides a foundation for implementing hierarchical reinforcement learning algorithms in traffic control systems. The framework establishes a two-tier agent hierarchy where Sector agents manage groups of Junction agents, enabling coordinated traffic optimization across urban networks.\n\n## Architecture\n\n### Agent Types\n\n- **Junction Agents**: Represent individual traffic intersections\n  - Subscribe to sector agents for coordination\n  - Receive state updates from external traffic simulation\n  - Maintain local traffic state information (coordination, priority, traffic counts)\n\n- **Sector Agents**: Manage groups of junction agents\n  - Handle junction subscription requests\n  - Negotiate with other sectors for junction assignments\n  - Coordinate junction behaviors within their sector\n\n### Key Components\n\n- **PlatformMediator**: Central management interface for the JADE platform\n  - Manages agent lifecycle (creation, initialization, updates)\n  - Bridges between external systems and the agent platform\n\n- **ControllerNativeInterface**: JNI bridge for simulation integration\n  - Connects to external traffic simulation (e.g., PARAMICS)\n  - Enables real-time state updates from simulation environment\n\n### Communication Protocols\n\nThe system implements FIPA-compliant protocols:\n- **Subscribe/Inform**: Junctions subscribe to sectors for coordination\n- **Propose**: Sectors negotiate junction assignments with each other\n\n## Features\n\n- **Dynamic Agent Discovery**: Agents discover each other via JADE's Directory Facilitator\n- **Hierarchical Organization**: Two-tier structure for scalable traffic management\n- **External Integration**: JNI interface for connecting to traffic simulators\n- **Asynchronous Updates**: Object-to-Agent (O2A) communication for real-time state updates\n- **Distributed Negotiation**: Sectors autonomously negotiate junction assignments\n\n## Current Implementation Status\n\nThe framework currently provides:\n- ✅ Multi-agent communication infrastructure\n- ✅ Hierarchical agent organization\n- ✅ External simulation integration via JNI\n- ✅ Dynamic agent management\n- ✅ Basic negotiation protocols\n\nSupports simulation-based optimization with:\n\n- Reinforcement learning algorithms\n- Q-learning or policy gradient methods\n- Reward/cost functions for optimization\n- State-action space definitions\n- Learning rate and exploration strategies\n\n## Usage\n\n### Starting the Platform\n\n```java\n// Initialize platform with junction IDs\nPlatformMediator.startJadePlatform(\"101 102 103 104\");\n\n// Add sector agents\nPlatformMediator.initSectorAgent(\"S-001\");\nPlatformMediator.initSectorAgent(\"S-002\");\n```\n\n### Updating Junction States\n\n```java\n// Update junction with traffic data\nJunctionUpdateBean update = new JunctionUpdateBean(junctionID);\nupdate.incomingCounts = Arrays.asList(10, 15, 8, 12); // N,S,E,W\nupdate.priority = 45;\nupdate.coordination = 0; // E-W coordination\nPlatformMediator.updJunctionAgent(\"101\");\n```\n\n## Dependencies\n\n- JADE 4.3.1 or higher\n- Java 1.6 or higher\n- Native traffic simulation library (via JNI)\n\n## Future Development\n\nThis framework is designed to support hierarchical reinforcement learning algorithms such as:\n- Hierarchical Q-learning for sector-level coordination\n- Multi-agent actor-critic methods for junction control\n- Transfer learning between similar traffic patterns\n- Coordinated exploration strategies for system-wide optimization\n\n## License\n\n[License information to be added]\n\n## Contributing\n\n[Contribution guidelines to be added]\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcdvel%2Fhlea4tc","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcdvel%2Fhlea4tc","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcdvel%2Fhlea4tc/lists"}