Hironmoy Roy Rudra

certified terrain traversal, target IROS 2027 · first author

Drive only on certified ground: a distribution-free margin on step, slope and clearance, a rule for which depth sensor can certify which rover, and active perception when nothing ahead is certified.

ICRA BARN Challenge preparation

A classical Jackal stack for hidden obstacle worlds: occupancy and distance field, A*, elastic band, MPC as a QP at 30 Hz, safety shield and recovery, under a 500-trial campaign protocol.

European Rover Challenge build, Nov 2025 – present

Direct technical strategy for the European Rover Challenge build and mentor the autonomy sub-team on navigation, perception, ROS 2 architecture and sim-to-real validation.

Jul – Aug 2026

• ROS 2 Jazzy
• C++
• Clearpath Jackal
• 2-D LiDAR
• MPC as a QP at 30 Hz

Full classical autonomy stack for the ICRA BARN Challenge (Clearpath Jackal, 2-D LiDAR, hidden obstacle worlds): online occupancy and distance field, A*, elastic band, sequentially linearised MPC solved as a QP at 30 Hz, safety shield, recovery; ROS 2 Jazzy, C++; a 500-trial campaign protocol on the BARN 2026 public 50-world suite; a nine-chapter tutorial and a live site with seven interactive demos.

Sep 2026 – present

• ZED 2i
• elevation_mapping_cupy
• TOPP-RA
• cvxpy
• Isaac Sim 6.0.1
• ROS 2 Jazzy

Sim-first research stack behind CERTUS: ZED 2i stereo depth through elevation_mapping_cupy to slope, roughness and uncertainty layers; TOPP-RA (cvxpy) velocity profiles whose limits vary along the path; Isaac Sim 6.0.1 scenes for Clearpath Husky, Jackal and Warthog; a ROS 2 Jazzy container; and a staged, idempotent setup script that pins every dependency by checksum and fails unless it reproduces the reported run.

Sep 2026

• Fusion 360
• PyBullet
• Gazebo Classic
• ROS 2 Humble

A closed-loop rocker-differential rover taken from Fusion 360 into PyBullet and Gazebo Classic: the loop stays closed to 0.003 mm in PyBullet, dead points found at −21.5° and +40° of rocker travel with ±15° hard stops recommended, skid versus ±90° steering compared on Apollo 15 terrain.

2026

• C++17
• pybind11
• Hybrid A*
• OpenCV

Bidirectional multi-heuristic hybrid A* for Ackermann and differential-drive vehicles: parallel 2-D Dijkstra heuristics in C++17 via pybind11, clearance-aware path simplification and Bézier smoothing that respects the minimum turning radius.

Dec 2025 – present

• Unity
• MuJoCo
• ROS 2
• URDF

High-fidelity Unity twin of the BRACU Mongol-Tori rover with a MuJoCo physics backend, URDF joints and custom mesh collisions for the rocker-bogie suspension; tied to a ROS 2 workspace so autonomy edge cases are validated virtually before hardware.

Aug 2025

• LiDAR point clouds
• Artificial potential fields
• Python

Artificial-potential-field planner over raw LiDAR point clouds that estimates terrain slope and isolates high-risk zones for safe waypoint traversal on rough ground.

IEEE/RSJ IROS 2027, first author

A rover drives only on ground it has certified: step, slope and body clearance pass with a distribution-free margin on the map's own uncertainty; a decidability rule predicts which depth sensor can certify which rover; when nothing ahead is certified, active perception picks the viewpoint that resolves the most blocking cells per second. Validated in Isaac Sim before deployment on a real rover.

IEEE ICRA 2027, second author

14 h 35 min across 175 scenes from a Livox Mid-360, stereo depth, six RGB cameras and post-processed GNSS/INS, with CNG auto-rickshaws and cycle rickshaws as 3D classes and every metric reported per traffic-density level.

IROS 2027, first author

The first collaborative-perception system whose shared representation is event-derived, with a counterfactually supervised, recipient-specific gate deciding what crosses the link; simulation study in Unreal Engine 5.8 / Cosys-AirSim.

NeurIPS 2027 Datasets & Benchmarks track, collaboration

Five industrial environment families with seeded layouts, exact label geometry, QC gates and rule-breaking hazard speeds, captured with event and RGB cameras.