forensicR

Tools for crime scene investigators, built with one.

The core is shooting incident reconstruction with explicit uncertainty, following the standard field workflow (Haag & Haag 2020; Hueske 2015; OSAC 2025-N-0003 terminology):

• impact angle from the defect ellipse, with a Monte Carlo interval;
• trajectories from rod/laser angles, from two points, or from a single defect's ellipse plus directionality;
• possible origin of fire at assumed muzzle heights, bounded by scene size;
• segmented paths through intermediate targets, with measured deflection angles and joint-consistency checks, or explicitly flagged assumed deflections when the path after the object was not documented;
• convergence of several trajectories (closest approach, miss distance, whether it lies behind both defects);
• plan and elevation plots showing the cone of uncertainty, not just a line.

Supporting modules:

• Scene mapping: baseline, triangulation and polar/total-station measurements to one XY frame (plus height z), plan view with room outline, and wall elevation views.
• Scene objects: furniture, appliances and persons as schematic boxes placed the way they are measured (against a wall, by a corner, by two corners), semi-transparent by default, drawn in every view; trajectories are checked for objects on their back-projected path.
• 3D reconstruction: the same objects rendered as a schematic, to-scale room with numbered floor axes, evidence markers, bullet defects at height, trajectory rods and uncertainty cones. A rayrender still for PDF/Word, an interactive rgl widget for HTML (both optional dependencies).
• Evidence documentation: evidence items, chain of custody, photo hashes, and a structured death scene observation record (observations only; time since death is the coroner's determination).

Reports render to Word, PDF and HTML from one template, each ending with a one-line note of the forensicR version that produced it.

Installation

# install.packages("pak")
pak::pak("kaquadros/forensicR")

Quick start

library(forensicR)

pts <- coords_baseline(along = c(2.4, 5.1), offset = c(1.2, -0.8), id = c("1", "2"))
plot_scene(pts)

impact_angle(width = 8, length = 16, se = 0.5)
t1 <- trajectory_from_angles(c(2.0, 5.0, 1.35), azimuth_deg = 352, vertical_deg = -3, id = "D2")
t2 <- trajectory_2pt(c(3.4, 4.2, 1.20), c(3.73, 4.92, 1.10), se_position = 0.01, id = "D3")
origin_zone(t1, heights = c(standing = 1.5), max_distance = 12)
intersect_trajectories(t1, t2)$point
plot_trajectories(list(t1, t2), view = "plan")

log <- evidence_log("2026-001234", "Lawrence Police Department")
log <- log_item(log, "1", "Cartridge case, 9 mm", location = "1", collected_by = "CSI")

render_scene_report(output_dir = "report")   # docx, pdf and html