Skip to main content
Formal Specification

Mauritanian camel caravan race and mimicry game — Formal Specification

Mauritania · festival-ritual · Confidence: medium

← Back to game page

Implementation Suitability

Suitability: engine-readyFidelity: discretizedGap-fills: 0
# Tmekter el-Jimal — AI-optimized specification ## General idea A multi-agent snake-chain navigation and synchronization race. A lead Guide agent steers a linked file of (K−1) Camel follower agents across a 2D desert track with checkpoints, executing vocal speed, turning, and synchronized kneeling commands (*Baraka*) while maintaining link connectivity and cargo balance. ## Formal model - **Abstraction taken:** The caravan is formalized as a kinematic snake chain of K nodes connected by distance constraints di,i+1∈[1.0,1.8]m. Synchronized commands require follower response latencies τi≤τthreshold. Cargo balance is an integrity state Ccargo∈[0,100]. - **Player/agent count:** 2 competing teams (TA,TB), each with K=4 agents (1 Guide + 3 Camels). - **Turn model:** Synchronous real-time ticks (Δt=100ms). ## Formal rules R1. **Track Geometry & Setup:** - Track X∈[0,60]m, width Y∈[−5,5]m. - Start at X=0; Checkpoint 1 at X=20m (Kneeling/Watering zone); Finish line at X=60m. - Each team caravan starts at X=0, link integrity = 100, cargo integrity = 100. R2. **Guide Steering & Vocal Broadcast:** - Guide agent A0 steers path (Vx,Vy) and issues discrete vocal commands: - `command_trot()`: Accelerate chain forward. - `command_turn(angle)`: Initiate steering curve. - `command_kneel()`: Order simultaneous transition to kneeling state (V→0, Z→0). - `command_rise()`: Order simultaneous rising from kneeling state. R3. **Follower Chain Physics & Synchronization:** - Follower agent Ai must track node Ai−1 coordinate. - If distance ‖Pi−1−Pi‖>2.0m: - Chain breaks (`broken_link = true`). - Speed of entire caravan drops to 0 until follower reconnects (‖Pi−1−Pi‖≤1.2m). - On `command_kneel()`: All followers must execute `kneel()` within timing tolerance Δt≤3 ticks (300ms). - If max⁡(Δt)≤3 ticks: Flawless synchronization (+2 bonus pts). - If any follower drops cargo: Cargo penalty applied. R4. **Scoring & Race Resolution:** - Total Score = Crossing Time Rank Points + Synchronization Bonuses − Link Break Penalties. - The team with higher aggregate score upon completing the 60m course wins. ## Edge cases - **Tie at the end condition:** Broken by higher remaining cargo integrity; if equal, broken by lowest link-break count. - **A player with no legal move:** Movement along track or executing vocal commands is always legal. - **Repetition / cycles:** Bounded by monotonic track advancement toward X=60m and hard tick cap Tmax⁡=1000 ticks (100 seconds). - **Simultaneous claims on the same resource or position:** Parallel team lanes prevent spatial collision between rival caravans. - **Player elimination mid-round; turn order repair:** If a follower agent disconnects, the chain enters recovery mode until repaired. - **Resource exhaustion:** Bounded track length (60m). - **Termination guarantee:** Guaranteed by finite track distance and monotonic progress of the Guide agent. ## State representation - **Persistent state:** Match scores (SA,SB), current checkpoint index. - **Caravan state:** Guide command state, link status boolean, cargo integrity Ci∈[0,100]. - **Agent state:** Position (Xi,Yi), velocity, stance ∈{standing,trotting,kneeling}, follower latency τi. - **Observable information:** Fully observable state within each caravan; guide broadcasts commands publicly to all team followers. ## Action space - For Guide: `steer(dir_x, dir_y, speed)`, `broadcast_command(cmd_type)`. - For Follower Camels: `follow_path(dx, dy)`, `execute_stance(stance_type)`. ## Playability notes - **Single-player-vs-AI:** Highly original: human plays as Guide (navigating turns and timing vocal commands) leading AI follower camels, or as lead camel matching guide cadence against an AI caravan. - **Group games (K=4):** Multi-agent cooperative physics benchmark (snake-chain formation). - **Physical-skill games:** Snake-chain physics and command latency modeled via spring-damper constraints. - **What is lost in the adaptation:** The physical laughter of stumbling into soft sand and singing camel herding chants. ## Implementation hints - Implement caravan followers using a standard distance-constraint inverse kinematics (IK) / Verlet rope solver. - First milestone: Implement 3-agent linked chain following a 2D guide vector without link breakage before adding kneeling commands.

Download as JSON

Download the full game corpus entry for this game as structured JSON.

Download JSON
Cite this page

Mauritanian camel caravan race and mimicry game (Mauritania). (n.d.). Games of Countries. https://gamesofcountries.com/games/mr-mauritania/tmekter-el-jimal

@misc{gamesofcountries-mr-mauritania-tmekter-el-jimal,
  title = {Mauritanian camel caravan race and mimicry game (Mauritania)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/mr-mauritania/tmekter-el-jimal}
}