[ SYSTEM: READY ]
[ GRAD: MAY 2026 ]
[ CORE: V4.0.2-STABLE ]
[ LATENCY: 12ms ]

MS Robotics & Autonomous Systems · ASU

Robotics
Engineer

Saiteja Dasari

M.S. Robotics candidate with 6+ projects across autonomy, controls, perception, and UAV systems. Experience deploying real-time robotic stacks from MATLAB and Simulink digital twins to ROS2 hardware, including multi-robot coordination, fault-tolerant flight, and vision-based landing. Strong in Python, C/C++, MATLAB/Simulink, and Linux.

Target Roles

Robotics EngineerControls EngineerUAV SystemsSystems EngineerRobotics Software EngineerAutonomy EngineerTest & Validation Engineer
bash : 80x24
01saiteja@asu:~$ros2 launch autonomy bringup.launch.py
02[INFO] [launch]: Starting /fdi_controller
03[INFO] [launch]: Starting /path_planner_node
04saiteja@asu:~$ros2 topic echo /system_status
05{
06 "identity": "Saiteja Dasari",
07 "gpa": "3.89 / 4.0",
08 "focus": ["Controls", "Vision", "Autonomy"],
09 "status": "NOMINAL"
10}
11saiteja@asu:~$
[ SYSTEM: READY ][ MODE: BIO ]ORIGIN_LOG

Origin Log

Operator

Saiteja
Dasari

Base of OperationsTempe, AZ
StatusActive

Bio

M.S. Robotics & Autonomous Systems candidate at Arizona State, graduating May 2026, with work spanning autonomy, controls, perception, fault-tolerant flight, and UAV systems engineering. I build robotic systems that move from digital twin to flight hardware, and I am targeting full-time robotics, controls, autonomy, and UAV roles.

Recent work includes deriving and hardware-validating fault-tolerant control for the Crazyflie 2.1 (max roll cut from 145° to 7.2°), centralized multi-robot task allocation across 5 TurtleBots in ROS2, vision-based autonomous landing on a moving platform at 8 cm accuracy, real-time YOLOv8 crop and weed detection at 12 FPS on ROSMaster X3, and the voice-analytics pipeline for an LLM-agent ATC simulator at Honeywell Aerospace.

I work across MATLAB and Simulink, Python, C and C++, ROS1 and ROS2, Gazebo, MuJoCo, CrazySim SITL, ArduPilot, and Linux. I care about clean abstractions, repeatable tests, and control loops that hold up when sensors lie.

Primary Focus

Robot AutonomyControlsMulti-Robot CoordinationFault-Tolerant FlightReal-Time PerceptionSim-to-Real Validation
[ SYSTEM: READY ][ LATENCY: 12ms ][ MODE: SPECS ]v4.0.2_CORE

Technical Specification

01code

Programming & Tools

Day-to-day stack for robotics software, simulation, and analysis. Python and C/C++ across embedded, ROS2 nodes, and tooling. MATLAB/Simulink for control design and digital twins.

LanguagesPython, C/C++
PythonC/C++MATLAB/SimulinkBashLinuxGitSolidWorks
02auto_videocam

Robotics & Simulation

Centralized ROS2 autonomy and Gazebo multi-robot environments. MuJoCo and CrazySim SITL for higher-fidelity quadrotor dynamics, with ArduPilot and Mission Planner on production UAVs.

Sim StackGazebo + MuJoCo
ROS1/ROS2GazeboMuJoCoCrazySim SITLRViztf2URDF/xacroArduPilotMission Plannercflib
03settings_input_component

Controls, GNC & Test

Fault detection and isolation, control allocation, and controller reconfiguration on Crazyflie 2.1. SITL workflows, log replay, and flight-log analysis on hardware.

Roll Reduction145° → 7.2°
PIDClosed-Loop ControlState EstimationSensor FusionFDIControl AllocationController ReconfigurationSITLLog ReplayFlight-Log Analysis
04visibility

Autonomy & Perception

Hungarian task assignment and A* planning over costmaps for multi-robot fleets. YOLOv8 and OpenCV pipelines for weed detection; ArUco-based pose estimation for the maze solver arm task.

Detection97% @ 12 FPS
A*Hungarian AlgorithmPath PlanningCostmapsOccupancy GridsCollision AvoidanceOpenCVYOLOv8ArUcoPose Estimation
05flight

Flight Platforms

Hands-on with Crazyflie 2.1 (fault-tolerant control), Parrot Mambo (vision-based landing), and P80 heavy-payload multirotor (ArduPilot PID tuning across 3 variants).

Landing CEP8 cm
Crazyflie 2.1Parrot MamboP80 MultirotorArduPilotMission Planner
06engineering

Sim-to-Hardware

Digital twin in Simulink for autonomous landing. MuJoCo rigid-body sim for gyroscopic coupling on Crazyflie, with CrazySim SITL bridging to firmware-in-the-loop testing.

Sim Conditions24 validated
Digital TwinMuJoCoCrazySim SITLSimulinkGazebo

Engineering Meta Data

Revision: STABLE-4.0
ParameterValueVarianceStatus
Max Roll Reduction145° → 7.2° via thrust clamp on Crazyflie 2.1± 0.5°[ VALIDATED ]
Allocation Methods Tested9 methods across 4 single-motor failure casesn/a[ ACTIVE ]
Sim Attitude Error0° across 24 test conditions in MuJoCon/a[ NOMINAL ]
CV Detection Accuracy97% @ 12 FPS, YOLOv8n on ROSMaster X3± 0.5%[ PEAK ]
Multi-Robot Success Rate99% across 5 TurtleBots, 10 Gazebo scenarios0.01%[ OPTIMIZED ]
Autonomous Landing CEP8 cm radius, Parrot Mambo, moving platform± 1.2 cm[ VALIDATED ]
UAV Stability Gain10% improvement, ArduPilot PID, 3 UAV variants± 1%[ NOMINAL ]
[ SYSTEM: READY ] [ MODE: BUILD_LOG ]

Build Log

Experience Record: 2023 to Present

01Aug 2025 to PresentACTIVE
Externship

Research Externship / Technical Lead, SkySpeak AI

Honeywell Aerospace, Arizona State University Collaborative Research

Directing a 3-engineer team in partnership with Honeywell and ASU stakeholders to deliver an ATC-pilot simulator built on LLM agents, structured guardrails, and a 300 ms LiveKit, Deepgram, and ElevenLabs voice pipeline. Engineered the Python voice analytics stack with YIN pitch detection, voice activity detection, z-score calibration, and abstention scoring under 0.60. Launched a CBTA dashboard used in 20+ stakeholder reviews and seed-grant evaluations.

LLM AgentsLiveKitDeepgramElevenLabsYINVADCBTAPythonReact
VISIT LIVE PRODUCTopen_in_new
02Jan 2025 to PresentACTIVE
Academic

Graduate Teaching Assistant

Arizona State University, MATLAB Programming and Instrumentation & Controls Lab

Mentoring 90+ students across MATLAB, instrumentation, and controls labs at ASU's Fulton Schools of Engineering. Troubleshooting code and hardware setups, leading review sessions, and reinforcing PID and control-system concepts during weekly office hours.

MATLABInstrumentationControlsPIDLab Instruction
03Oct 2023 to Mar 2024COMPLETED
Industry

Production Engineering Intern, UAV Systems Integration

Marut Drones (IIIT-Hyderabad)

Assembled and verified UAV subsystems including ESCs, flight controllers, GPS, avionics, and propulsion hardware across multirotor platforms, raising production efficiency by 15%. Tuned ArduPilot and Mission Planner PID parameters from flight logs across 3 UAV variants, root-caused stability issues, and increased flight stability by approximately 10%.

ArduPilotMission PlannerPIDUAV AssemblyFlight Logs3 Variants
04Feb 2026 to PresentACTIVE
Startup

Co-Founder

Aatram

Co-founded an emotion-first anti-procrastination app for students and young professionals. Designed an adaptive nudge engine leveraging Apple Intelligence for context-aware notifications. Built a Momentum Board with streak alternatives (Bounce-Back Score, Monthly Chapters, momentum tracking), replacing punitive streak systems with psychologically grounded progress mechanics. Integrated evidence-based focus techniques: implementation intentions, WOOP, affect labeling, and temptation bundling.

Product ArchitectureJITAIBehavioral DesignCloudflare Workers
VISIT LIVE PRODUCTopen_in_new
[ FEED: LIVE ][ MODE: NOW ]SYS_TIME: APR 2026
[ NOW: APR 2026 ]
current_focus
  • >Validating MuJoCo to hardware sim-to-real on Crazyflie 2.1
  • >Extending SkySpeak AI for live demo and seed-grant review
  • >Interviewing for full-time robotics, controls, and UAV roles
  • >Graduating from ASU in May 2026
>
[ SYSTEM: READY ] [ MODE: DEPLOYMENTS ]

Deployments

Project Archive: 2024 to 2026

001

Fault-Tolerant Quadrotor Control

Jan 2026 to Present

Max Roll Reduction

145° to 7.2°

Sim Conditions

24 validated

Allocation Methods

9 tested

View Full Detailarrow_forwardterminalRepoopen_in_new
Open Detail View
allocator.pycode
# 3-motor allocation,
# motor 0 failed
W = clamp_thrust(
    M_3motor @ cmd_wrench
)

motor.set_pwm(
    [0.0, *W]
)

Closed-form 3-motor allocation derived and validated for all 4 single-motor failure cases. Identified motor saturation as the root cause of attitude loss; a thrust clamp reduced max roll from 145° to 7.2° on hardware. MuJoCo rigid-body sim built for gyroscopic coupling, with sim-to-real validation across PID, INDI, and gyroscopic-compensation controllers.

002

Multi-Robot Task Allocation

Dec 2025

Robots

5 TurtleBots

Success Rate

99%

Scenarios

10 validated

View Full Detailarrow_forwardterminalRepoopen_in_new
Open Detail View
task_allocator.pyschema
def allocate(robots, tasks):
    C = build_cost_matrix(
        robots, tasks
    )
    assign = hungarian(C)

    for r, t in assign:
        plan_path(r, t, costmap)

Centralized ROS2 autonomy stack in Gazebo for 5 TurtleBots. Hungarian task assignment and A* motion planning with costmap-based collision avoidance. Validated across 10 scenarios with 99% mission success.

003

Crop Weed Detection

May 2025

Accuracy

97%

Throughput

12 FPS

Dataset

3000 samples

View Full Detailarrow_forward
Open Detail View
weed_detector.pyfingerprint
def detect(self, frame):
    results = model.predict(
        frame, conf=0.5
    )
    for box in results.boxes:
        cls = box.cls.item()
        if cls == WEED_ID:
            self.flag(box.xyxy)

YOLOv8 model trained on 3000 samples, deployed on ROSMaster X3 at 97% accuracy and 12 FPS. Integrated ROS inference and visualization nodes for real-time robotic decision-making in precision agriculture.

004

Autonomous Drone Landing

May 2025

Landing Accuracy

8 cm CEP

Sim Trials

25 validated

Platform

Parrot Mambo

View Full Detailarrow_forwardterminalRepoopen_in_new
Open Detail View
landing_controller.pyflight
def track_target(self, frame):
    target = self.detect_marker(frame)
    error = target.pos - self.uav.pos

    cmd = self.pid.compute(error)
    self.publish_velocity(cmd)

Guidance and control logic built in MATLAB and Simulink as a digital twin across 25 trials. Onboard vision-based relative pose estimation on the Parrot Mambo, paired with closed-loop PID, achieves landing within 8 cm of a moving platform.

005

Maze Solving with myCobot600

Dec 2024

Grid

8×8

Plan Time

60 s

Arm

6-DOF myCobot600

View Full Detailarrow_forwardterminalRepoopen_in_new
Open Detail View
maze_solver.pyroute
grid = occupancy_grid(
    frame, aruco_markers
)
path = astar(grid, start, goal)

for waypoint in path:
    arm.move_to_waypoint(waypoint)

OpenCV pipeline for maze detection using 2 ArUco markers on an 8×8 occupancy grid. A* path computed and executed via myCobot600 6-DOF motion control; 60 s planning time. Applicable to industrial pick-and-place workflows.

006

SkySpeak AI

Aug 2025 to Present

Voice Pipeline

300 ms latency

Stakeholder Reviews

20+

Team Size

3 engineers

View Full Detailarrow_forwardboltLive Demoopen_in_newterminalRepoopen_in_new
Open Detail View
voice_analytics.pymic
frame = audio.next()
pitch = yin(frame)
score = z_score(
    pitch, baseline
)
if abstention(score) < 0.6:
    grade = cbta(transcript)
    dashboard.push(grade)

AI-powered ATC-pilot simulator for Honeywell Aerospace's Anthem ecosystem. Real-time voice pipeline at 300 ms (LiveKit, Deepgram, ElevenLabs) paired with Python voice analytics: YIN pitch, VAD, z-score calibration, and abstention scoring under 0.60. CBTA dashboard surfaces cognitive load and communication competency to instructors.

[ SYSTEM: READY ] [ MODE: MANIFEST ]

Manifest

Certifications, Publications & Education

Certifications

precision_manufacturing

Universal Robots e-Series

01

Core & Pro

Universal Robots

model_training

MathWorks AI Certification

02

Machine Learning & Deep Learning

MathWorks

psychology

Hugging Face RL Course

03

Reinforcement Learning

Hugging Face

smart_toy

ROS & ROS2

04

Robot Operating System

Udemy

Publications

article

SCADA-Based Substation Control Panel & Operations

open_in_newPUB-01

Technical Publication

Published Research

Education

Arizona State University

M.S. Robotics & Autonomous Systems, Systems Engineering Track

GPA: 3.89 / 4.0

Coursework: Aerial Robotics, Mechatronics, Multi-Robot Systems, Controls & Systems

2024 to 2026

Chaitanya Bharathi Institute of Technology

B.E. Electrical & Electronics Engineering

Hyderabad, Telangana

2020 to 2024
[ SYSTEM: READY ][ MODE: CONTACT ]OPEN FOR OPPORTUNITIES

Initialize Contact

Available for full-time roles, internship roles, collaboration, and research roles

Graduating

May 2026

Open For

Full-time Robotics, Controls, Autonomy, UAV, Test & Validation roles

Location

Tempe, AZ · Open to relocation

Schedule a CalleventBook on Calendly
personLINKEDIN PROFILEterminalGITHUB REPOS

Current Status

Open for Opportunities