traffic op• Data kinds: none → table (an op determined by its arguments alone — it takes no image or data input)
• Call: import fullseye as fs; fs.ledger.pedestrian_crossing(intent: 'str', *, start_xy=(0.0, -3.0), road_width: 'float' = 7.0, speed: 'float' = 1.2, wait_time: 'float' = 3.0, curb_offset: 'float' = 0.3, along_heading: 'float' = 0.0, dt: 'float' = 0.1, t_end: 'Optional[float]' = None) -> 'Dict[str, object]' (to call the implementation directly, import drivetraffic; drivetraffic.pedestrian_crossing(intent: 'str', *, start_xy=(0.0, -3.0), road_width: 'float' = 7.0, speed: 'float' = 1.2, wait_time: 'float' = 3.0, curb_offset: 'float' = 0.3, along_heading: 'float' = 0.0, dt: 'float' = 0.1, t_end: 'Optional[float]' = None) -> 'Dict[str, object]'; from the registry, opsdrive.get("pedestrian_crossing"))
Trajectory of a pedestrian whose crossing intention is the ground truth (closed form per segment; no integration).
The road runs in the x direction, carriageway 0 ≤ y ≤ road_width, starting from the pavement y < 0 (y of start_xy < −curb_offset).
• cross: walks from (x0, y0) to the kerb (x0, −curb_offset), crosses without stopping to the far kerb (x0, road_width + curb_offset) and stops. Heading +π/2 (toward the carriageway).
• wait_then_cross: walks to the kerb, stops facing the carriageway at the kerb (+π/2) for wait_time seconds, then crosses.
• walk_along: walks along the pavement facing along_heading (0 or π) staying at y = y0 (does not enter the carriageway).
• stand_no_cross: walks to the kerb and keeps standing facing the carriageway at the kerb (+π/2) without crossing (waiting for a bus or a companion). In appearance it cannot be distinguished from the wait of wait_then_cross — a ground truth that becomes a false detection if a rule judges "stopping" to be the cue. t_cross_start is None.
Speed is speed (default WALK_SPEED = 1.2 m/s, assumed).
Returns dict: t (K,), xy (K, 2), heading (K,), phase (K,) strings ("approach" / "wait" / "cross" / "done" / "along"), intent (ground truth), t_cross_start and t_cross_end (None for intents that do not cross), speed. Gate: crossing time = (road_width + 2·curb_offset)/speed; a waiting person has constant position and heading +π/2 during the wait; a person walking along has y < 0 at all times; the heading matches the direction of the position differences.
Raises ValueError: unknown intent, speed ≤ 0, start not on the pavement, along_heading not 0/π.
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• poc_driving_traffic — py -3.11 examples/poc_driving_traffic.py
table as input)course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same
traffic)idm_accel · idm_equilibrium_gap · idm_platoon_simulate · driver_style · lateral_wobble · ou_estimate · social_force_step · occlusion_reveal_distance
*Provenance: drivetraffic.py — DRIVE operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.