Traffic Policies¶

Policies governing robot movement and congestion.

Overview¶

Traffic policies control how robots navigate, handle congestion, and resolve conflicts.

Policy Area	Decisions
Routing	Which path to take
Congestion	Wait or reroute
Deadlock	How to resolve
Priority	Who goes first

Routing Policies¶

Shortest Path¶

Always take minimum distance route:

routing:
  policy: shortest_path

A ──3── B ──2── C
│               │
4               3
│               │
D ──────5────── E

A to E: A→B→C→E (8m) ✓
Not:    A→D→E (9m)

Pros: Minimum travel distance Cons: May route through congestion

Congestion-Aware¶

Avoid congested areas:

routing:
  policy: congestion_aware
  congestion_weight: 1.5

Cost calculation:

path_cost = distance + congestion_weight × congestion_score

Path 1: 8m, congestion=5  → cost = 8 + 1.5×5 = 15.5
Path 2: 10m, congestion=1 → cost = 10 + 1.5×1 = 11.5 ✓

Pros: Avoids delays Cons: Longer distances, requires congestion data

Time-Optimal¶

Minimize expected travel time:

routing:
  policy: time_optimal

Considers: - Distance - Current congestion - Historical patterns - Expected wait times

Congestion Response¶

Wait¶

Robot waits when path blocked:

traffic:
  congestion_response: wait

R1 → [blocked] → destination

R1 waits at current position

Pros: Simple, no replanning Cons: May wait long time

Reroute¶

Find alternative path:

traffic:
  congestion_response: reroute
  reroute_threshold_s: 5.0

R1 → [blocked] → destination
          ↓
R1 → [alternative path] → destination

Pros: Avoids long waits Cons: Longer distance, replanning overhead

Hybrid¶

Combine strategies:

traffic:
  congestion_response: hybrid
  wait_threshold_s: 3.0
  reroute_threshold_s: 10.0

Wait < 3s:  Wait
3s-10s:     Continue waiting
Wait > 10s: Reroute

Deadlock Policies¶

Priority Resolution¶

Higher priority robot proceeds:

traffic:
  deadlock_resolution: priority

R1 (priority 7) ←→ R2 (priority 3)

R2 backs off, R1 proceeds

Time-Based Resolution¶

Robot waiting longer proceeds:

traffic:
  deadlock_resolution: time_based

R1 waiting 5s ←→ R2 waiting 2s

R1 proceeds (waited longer)

Backoff Resolution¶

One robot retreats:

traffic:
  deadlock_resolution: backoff
  backoff_selection: random

Options: random, lower_priority, further_from_goal

Traffic Flow Policies¶

One-Way Aisles¶

Enforce directional flow:

traffic:
  one_way_enforcement: true
  aisle_direction:
    aisle_1: forward  # →
    aisle_2: backward # ←

Aisle 1: →→→→→→→→→
Aisle 2: ←←←←←←←←←

Lane Discipline¶

Multiple lanes with rules:

traffic:
  lanes:
    main_aisle:
      lanes: 2
      policy: keep_right

Lane 1: →→→→→ (rightward traffic)
Lane 2: ←←←←← (leftward traffic)

Speed Zones¶

Variable speed by area:

traffic:
  speed_zones:
    main_aisle:
      max_speed_m_s: 2.0
    intersection:
      max_speed_m_s: 0.5

Priority in Traffic¶

Right-of-Way¶

Higher priority robots have precedence:

traffic:
  right_of_way: priority_based

Intersection conflict:
  R1 (priority 7) vs R2 (priority 3)
  → R1 goes first

Emergency Priority¶

Emergency tasks override all:

traffic:
  emergency_priority: true
  emergency_threshold: 9

Priority 9+ robots: - Skip queues - Others yield immediately - Reroute lower priority robots

Configuration Example¶

routing:
  policy: congestion_aware
  congestion_weight: 1.5
  replan_on_congestion: true
  replan_threshold_s: 10.0

traffic:
  # Congestion response
  congestion_response: hybrid
  wait_threshold_s: 3.0
  reroute_threshold_s: 15.0

  # Deadlock handling
  deadlock_detection: true
  deadlock_resolution: priority
  deadlock_timeout_s: 30.0

  # Flow control
  one_way_enforcement: false
  right_of_way: priority_based

  # Speed
  default_speed_m_s: 1.5
  intersection_speed_m_s: 0.8

Policy Interactions¶

Routing + Congestion¶

Congestion-aware routing → Fewer blocks → Less rerouting

Priority + Deadlock¶

Priority routing → Fewer conflicts → Simpler deadlock resolution

Batching + Traffic¶

Batching → Fewer trips → Less congestion → Better throughput

Comparing Policies¶

Simulation Comparison¶

waremax compare scenario.yaml \
  --param routing.policy=shortest_path \
  --param routing.policy=congestion_aware

Metrics to Watch¶

Metric	Indicates
Travel time	Route efficiency
Wait time	Congestion handling
Deadlocks	Resolution effectiveness
Throughput	Overall efficiency

Best Practices¶

Start Simple¶

Begin with basic policies:

routing:
  policy: shortest_path

traffic:
  congestion_response: wait

Add Complexity as Needed¶

If congestion is a problem:

routing:
  policy: congestion_aware

If deadlocks occur:

traffic:
  deadlock_detection: true
  deadlock_resolution: priority

Monitor and Tune¶

Track key metrics and adjust:

High wait time → Enable rerouting
Frequent deadlocks → Improve prevention
Uneven flow → Consider one-way aisles