The positioning stack behind the products

Proprietary GNSS silicon, an in-house RTK/PPP engine, global corrections and sensor fusion with vision and LiDAR — developed at Tersus since 2014.

Signal to deliverable

Antares™ SoC → ExtremeRTK™ engine → TAP™ PPP corrections → visionRTK™ / TightSLAM™ fusion → capturRTK™ reconstruction

Signal chain from satellite and SoC through RTK, PPP, scanner fusion and office reconstruction

Antares™

GNSS SoC

TAP™ PPP

GNSS constellation · L-band / IP corrections

ExtremeRTK™

RTK engine

visionRTK™

Visual-inertial positioning

TightSLAM™

GNSS/IMU-anchored SLAM

capturRTK™

Georeferenced reconstruction

MVP Engine

Office reconstruction

Red markers show where each technology lives in the hardware and software chain.

Technologies

Antares™

Proprietary GNSS SoC

Tersus designs its own GNSS system-on-chip: signal tracking, multipath mitigation and the RTK engine run on silicon Tersus controls, which is why receiver performance can be tuned for high-accuracy surveying applications and for machine control in challenging environments.

Schematic of a patch antenna feeding three GNSS signal traces into a single system-on-chip, which outputs one computed position marked in red.

ExtremeRTK™

Positioning engine

Multi-constellation, multi-frequency RTK with fast re-acquisition under canopy and near structures. Patented work in signal tracking, multipath mitigation and GNSS/IMU fusion.

Schematic of a survey rover between a tree canopy and a building: two satellite signals arrive clear, two are blocked, and a dashed correction link from a base station keeps the fixed position marked in red.

TAP™ PPP

Global corrections

Satellite- and IP-delivered PPP corrections, no base station and no radio link. Converges to under 3 cm horizontal in under 5 minutes under open sky (95 % confidence).

Schematic of a survey rover alone in an open field receiving satellite signals and a dashed IP correction stream from a cloud, with a spiral converging on the receiver position; no base station.

visionRTK™

Visual-inertial positioning

Tersus has pioneered the integration of camera-based visual odometry, visual SLAM and GNSS/INS on a survey rover to maintain a usable position where satellites drop out, under canopy, beside structures, in urban canyons.

Schematic of a survey rover with a camera module projecting a frustum onto a building facade, tracked visual features marked as red dots and one satellite link to the antenna.

TightSLAM™

GNSS/IMU-anchored SLAM

Fuses GNSS and IMU data directly into the SLAM pipeline, anchoring 3D spatial data to real-world coordinates instead of estimating it from imagery alone, so mapping stays survey-accurate underground and between high-rise buildings.

Schematic of a handheld lidar scanner inside a tunnel with scan lines fanning to the walls; its trajectory is anchored by a GNSS antenna at the entrance and red control points.

capturRTK™

Georeferenced reconstruction

On-device mesh and point-cloud generation with RTK constraints, producing georeferenced deliverables rather than a floating scan.

Schematic of a handheld scanner projecting a scan cone onto a building corner, half of which is resolved as a triangulated mesh with red georeferencing anchor points.

MVP Engine

Simultaneous localisation and mapping

Real-time LiDAR and visual SLAM engine that builds dense, georeferenced 3D maps while the operator walks. Fuses IMU, GNSS and camera data for centimeter-level accuracy indoors and outdoors.

Schematic of an operator walking with a handheld scanner and GNSS antenna; the environment behind resolves into a dotted point-cloud map while scan lines fan ahead.

Need the engineering detail?

Read the documentation or talk to engineering.

Talk to engineering

Our Newsletter, right to the point

Product updates, case studies, tech articles and industry news.