FAILURE MAP
← Case archive

FA-93626 / Solar tracker geometry / Open access

Cross-axis slope aware backtracking: angle relative to slope · case 01

Rows backtrack on the downhill side when they should on the uphill side.

Verified by executionVariant 1 · 7 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The slope is added to the rotation instead of subtracted.

VERIFIED REPAIR

Evaluate cos(ideal - cross_tilt).

Unsuccessful approach: Ignoring the slope in the shading test backtracks as if the ground were flat.

Case contract

Rows on terrain with cross-axis slope cross_tilt degrees. dist = 1/(gcr*cos(cross_tilt)); temp = dist*cos(ideal - cross_tilt). When temp < 1 the angle becomes ideal - sign(ideal)*acos(max(temp, -1)) degrees, else ideal. The result is then clamped to +-max_angle. Return the angle rounded to 3.

Why this case matters

Single-axis and dual-axis solar trackers turn a sun direction into actuator commands; a sign, frame or limit mistake points a whole plant away from the sun or into a mechanical stop.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(ideal, gcr, cross_tilt, max_angle):
    dist = 1 / (gcr * math.cos(math.radians(cross_tilt)))
    temp = dist * math.cos(math.radians(ideal + cross_tilt))
    if temp < 1:
        theta = ideal - math.copysign(math.degrees(math.acos(max(temp, -1.0))), ideal)
    else:
        theta = ideal
    theta = max(-max_angle, min(max_angle, theta))
    return round(theta, 3)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [80, 0.3, -10, 55], -10.0],
  ['regression: angle relative to slope (partial repair)', [-60, 0.5, 5, 50], -28.045],
  ['control 1', [-45, 0.5, -5, 50], -45], ['control 2', [-60, 0.3, 0, 60], -60]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.5, -5, 55], 28.045],
  ['regression: angle relative to slope (partial repair)', [80, 0.4, 10, 60], 50.255],
  ['control 1', [80, 0.4, 0, 60], 15.729], ['control 2', [-80, 0.4, 15, 75], 23.037]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.4, 15, 55], 55],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 15, 60], 2.496],
  ['control 1', [0, 0.4, -5, 55], 0], ['control 2', [70, 0.4, -5, 75], 20.505]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-80, 0.4, -10, 75], -50.255],
  ['regression: angle relative to slope (partial repair)', [80, 0.3, -5, 55], 6.955],
  ['control 1', [-60, 0.4, -5, 50], -50], ['control 2', [0, 0.4, -5, 60], 0]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-70, 0.3, -5, 60], -60],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 5, 75], -40.0],
  ['control 1', [-80, 0.5, 0, 75], -10.322], ['control 2', [-45, 0.4, 15, 50], -45]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
normal: flat terrain matches plain backtracking-38.765-38.765Passed
boundary: slope-facing noon00Passed
boundary: clamp after backtracking50-10.0Failed
regression: angle relative to slope55-10.0Failed
regression: angle relative to slope (partial repair)-50-28.045Failed
control 1-45-45Passed
control 2-60-60Passed

SHA-256 / 64af543873ddbc7dff482c11dcd75df16eaef92d9d0856831a7c2be84c2f60f6

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(ideal, gcr, cross_tilt, max_angle):
    dist = 1 / (gcr * math.cos(math.radians(cross_tilt)))
    temp = dist * math.cos(math.radians(ideal))
    if temp < 1:
        theta = ideal - math.copysign(math.degrees(math.acos(max(temp, -1.0))), ideal)
    else:
        theta = ideal
    theta = max(-max_angle, min(max_angle, theta))
    return round(theta, 3)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [80, 0.3, -10, 55], -10.0],
  ['regression: angle relative to slope (partial repair)', [-60, 0.5, 5, 50], -28.045],
  ['control 1', [-45, 0.5, -5, 50], -45], ['control 2', [-60, 0.3, 0, 60], -60]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.5, -5, 55], 28.045],
  ['regression: angle relative to slope (partial repair)', [80, 0.4, 10, 60], 50.255],
  ['control 1', [80, 0.4, 0, 60], 15.729], ['control 2', [-80, 0.4, 15, 75], 23.037]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.4, 15, 55], 55],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 15, 60], 2.496],
  ['control 1', [0, 0.4, -5, 55], 0], ['control 2', [70, 0.4, -5, 75], 20.505]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-80, 0.4, -10, 75], -50.255],
  ['regression: angle relative to slope (partial repair)', [80, 0.3, -5, 55], 6.955],
  ['control 1', [-60, 0.4, -5, 50], -50], ['control 2', [0, 0.4, -5, 60], 0]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-70, 0.3, -5, 60], -60],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 5, 75], -40.0],
  ['control 1', [-80, 0.5, 0, 75], -10.322], ['control 2', [-45, 0.4, 15, 50], -45]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
normal: flat terrain matches plain backtracking-38.765-38.765Passed
boundary: slope-facing noon00Passed
boundary: clamp after backtracking25.998-10.0Failed
regression: angle relative to slope25.998-10.0Failed
regression: angle relative to slope (partial repair)-50-28.045Failed
control 1-45-45Passed
control 2-60-60Passed

SHA-256 / 8f50a3d09f28401db388771a5827fefd77814721a6b1f2661ae124207de002a6

3 / The verified repair

Exit 0
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(ideal, gcr, cross_tilt, max_angle):
    dist = 1 / (gcr * math.cos(math.radians(cross_tilt)))
    temp = dist * math.cos(math.radians(ideal - cross_tilt))
    if temp < 1:
        theta = ideal - math.copysign(math.degrees(math.acos(max(temp, -1.0))), ideal)
    else:
        theta = ideal
    theta = max(-max_angle, min(max_angle, theta))
    return round(theta, 3)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [80, 0.3, -10, 55], -10.0],
  ['regression: angle relative to slope (partial repair)', [-60, 0.5, 5, 50], -28.045],
  ['control 1', [-45, 0.5, -5, 50], -45], ['control 2', [-60, 0.3, 0, 60], -60]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.5, -5, 55], 28.045],
  ['regression: angle relative to slope (partial repair)', [80, 0.4, 10, 60], 50.255],
  ['control 1', [80, 0.4, 0, 60], 15.729], ['control 2', [-80, 0.4, 15, 75], 23.037]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [60, 0.4, 15, 55], 55],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 15, 60], 2.496],
  ['control 1', [0, 0.4, -5, 55], 0], ['control 2', [70, 0.4, -5, 75], 20.505]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-80, 0.4, -10, 75], -50.255],
  ['regression: angle relative to slope (partial repair)', [80, 0.3, -5, 55], 6.955],
  ['control 1', [-60, 0.4, -5, 50], -50], ['control 2', [0, 0.4, -5, 60], 0]],
 [['normal: flat terrain matches plain backtracking', [-70, 0.4, 0, 60], -38.765],
  ['boundary: slope-facing noon', [0, 0.4, 10, 60], 0],
  ['boundary: clamp after backtracking', [80, 0.3, -10, 50], -10.0],
  ['regression: angle relative to slope', [-70, 0.3, -5, 60], -60],
  ['regression: angle relative to slope (partial repair)', [-70, 0.3, 5, 75], -40.0],
  ['control 1', [-80, 0.5, 0, 75], -10.322], ['control 2', [-45, 0.4, 15, 50], -45]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
normal: flat terrain matches plain backtracking-38.765-38.765Passed
boundary: slope-facing noon00Passed
boundary: clamp after backtracking-10.0-10.0Passed
regression: angle relative to slope-10.0-10.0Passed
regression: angle relative to slope (partial repair)-28.045-28.045Passed
control 1-45-45Passed
control 2-60-60Passed

SHA-256 / 8378eb4b083aa738033d6125b7d6d8b4b2567b7080c74c10c6d48d0df1ddaa30

Verification & scope

Deterministic stipulated toy contract for teaching; no claim of conformance with any standard, vendor protocol or production controller. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:51:56.831553+00:00.

Case digest / 5c3d8e3f76feb7f4d7239d7c8941d6ed66adc83a31bfad1dbc267ac5d645dd5d