RESEARCH_MAP

Research Interests & Nodes

Interactive visualization representing the dependencies and intersections between autonomous system layers.

SYS_TOPOLOGY_VIEW [MOUSE_OVER_NODES]
Robotics IntegrationAutonomous NavigationComputer VisionMulti-Agent SystemsAI SystemsHuman Robot Interaction
NODE_DIAGNOSTIC_SUMMARY
ACTIVE_NODE_ID:
/robotics
DISPLAY_NAME:
Robotics Integration
CORE_OBJECTIVE:

Building unified platforms combining kinematics, real-time operating systems (RTOS), actuator controls, and high-level sensory computation.

TOPOLOGICAL_CONNECTIONS:
/navigation/cv/hri/ai
SYS_LIVE_TELEMETRY

Sensor Streams & Diagnostics

Interactive Hardware-in-the-Loop simulation interface. Switch system execution profiles to load real-time telemetry datasets.

FEED: /dev/lidar_serial
ACTIVE_DEVICE_CONFIG: SLAM NAVIGATION MODE
BAUD_RATE:
115200 bps
IMU_HEADING:
0.72 rad
LOCAL_POSE_X:
1.45 m
LOCAL_POSE_Y:
-0.89 m
NODE_LATENCY:
12 ms
GPU_UTILIZATION:
32 %
ARM_CPU_TEMP:
48.5 °C
BUS_VOLTAGE:
12.4 V
SUBSCRIBED_TOPICS:
/scan
@10HzACTIVE
/odom
@50HzACTIVE
/tf
@100HzACTIVE
/cmd_vel
@20HzACTIVE
ROUTING_STREAM: /var/log/ros_node.log
[INFO] [nav2_planner]: Planning path to coordinates: X=4.12, Y=-2.04
[INFO] [cartographer]: Scan-matching succeeded. Map updated.
[WARN] [ekf_node]: IMU variance exceeds limit, recalculating covariance.
[INFO] [nav2_controller]: Velocity commands regulated. Speed: 0.35m/s
[INFO] [nav2_planner]: Goal reached. System waiting for command.
AUTO_RELOAD: TRUE
TRANSLATIONAL_METHODOLOGY

Research translation pipeline

How theoretical equations are refined inside virtual environments, compiled into firmware, and deployed to physical hardware.

PHASE_01

Mathematical Theory

Kinematic formulations & stochastic sensor modeling.

PHASE_02

Simulation Sandbox

Gazebo physical dynamics & noise injection tests.

PHASE_03

Hardware-In-The-Loop

STM32 firmware compilation & Micro-ROS bridge.

PHASE_04

Field Deployment

Physical rover trials & real-time telemetry profiling.

PIPELINE_STATUS: COMPILED_SUCCESSFULLY
ACTIVE_STRETCH:/math_model/sim_sandbox/hil_mcu/jetson_edge
PHASE_SPEC_LOGS: #PHASE_01

Mathematical Theory

Formulating differential drive kinematics, EKF state estimators, and path optimization equations. We design obstacle costmaps and local planner trajectory constraints mathematically before writing control code.

ASSOCIATED_TECHNOLOGIES:
KinematicsState EstimationStochastic CalculusControl Theory
VALIDATION_CRITERIA_METRICS:
State Vector Dimensions6 (x, y, theta, vx, vy, vtheta)
Target Convergence Rate99.2%
PUBLICATIONS_&_REPORTS

Academic Preprints & Technical Notebooks

Publications translating core control theory and machine learning models into field-ready physical rovers and edge pipelines.

Robotics/Autonomous Robotics & Systems Journal (Preprint)/June 2026

A Hybrid SLAM and Predictive Path-Planning System for Differential Drive Rovers in Dynamic Warehouses

Authors: Aditya Sonwane

PREPRINT_PDFSOURCE_CODEDEMO_VIDEO
AI/CV/Journal of Real-Time Image Processing & AI (Technical Report)/May 2026

Edge-Inference Acceleration for Stereo-Vision Obstacle Detection on Embedded Jetson Platforms

Authors: Aditya Sonwane

PREPRINT_PDFSOURCE_CODEDEMO_VIDEO
Distributed Systems/International Conference on Multi-Agent Systems & Robotics (Preprint)/March 2026

Decentralized Swarm Navigation using Reciprocal Velocity Obstacles and Distributed Consensus Networks

Authors: Aditya Sonwane

PREPRINT_PDFSOURCE_CODEDEMO_VIDEO