
Completed
Posted
Paid on delivery
I’m running an ExternalNav pipeline that takes UWB data through a Raspberry Pi 5, packages it as MAVLink VISION_POSITION_ESTIMATE messages, and feeds it into ArduPilot. Somewhere along that path the position information degrades and the vehicle refuses to accept the estimates. I need your debugging skills to pinpoint the fault and deliver a clean, reliable flow of data all the way to ArduPilot. You’ll have full access to the C++ firmware on the UWB microcontroller and my current ArduPilot parameters. Mission Planner logs, real-time telemetry, and a reproducible bench test setup are ready to share the moment we start. Key technologies you’ll be working with include: • C++ (embedded STM32-class environment) • Python 3.x on Raspberry Pi 5 • MAVLink 2.0 message packing/parsing • ArduPilot 4.x EKF / ExternalNav configuration • Mission Planner for log inspection and parameter tuning • Decawave-based UWB anchors and tags Deliverables 1. A concise report explaining the root cause of the failure. 2. Updated code or configuration files that correct the issue. 3. A verification log (or short flight test) showing stable VISION_POSITION_ESTIMATE data accepted by ArduPilot.
Project ID: 40592709
31 proposals
Remote project
Active 6 days ago
Set your budget and timeframe
Get paid for your work
Outline your proposal
It's free to sign up and bid on jobs

The MAVLink timestamp drift often makes the EKF reject VISIONPOSITIONESTIMATE packets, which explains the degradation you see. I'll start by checking the time sync between the Raspberry Pi and the STM32, then verify the message packing against the MAVLink 2.0 spec. After that I'll adjust the ExternalNav parameters and push a corrected binary to the microcontroller for a clean data flow. A common mistake is sending the position estimate with a stale or out-of-range covariance, causing the EKF to drop the data. You’ll get a short report, updated source files, and a log showing ArduPilot accepting the vision messages without warnings.
$50 USD in 2 days
2.3
2.3
31 freelancers are bidding on average $47 USD for this job

Hi I am an embedded systems engineer with over 16 years of experience. I have worked on embedded C++ firmware, Raspberry Pi Python services, MAVLink telemetry, ArduPilot/PX4 integration, sensor fusion, and UWB-based positioning flows, so this ExternalNav issue is exactly the kind of fault I can debug methodically. My approach would be to trace the complete chain from Decawave UWB output to the STM32 firmware, Raspberry Pi MAVLink packing, message timing, coordinate frame, covariance, EKF source parameters, and Mission Planner logs. In many cases the issue is not just the message itself, but timestamping, origin/frame mismatch, estimator gating, message rate, covariance values, or EKF configuration rejecting otherwise valid data. I can review your C++ firmware, Python MAVLink bridge, current ArduPilot parameters, and logs, then return a concise root-cause report along with corrected code/configuration and a verification log showing ArduPilot accepting stable VISION_POSITION_ESTIMATE data. A few quick checks before starting: are you sending LOCAL_NED-compatible coordinates, what message rate is currently used, and do your logs show EKF innovation/gating messages when ExternalNav is enabled? Please contact me to discuss details.
$42 USD in 3 days
7.6
7.6

Hi, I have experience in designing Circuits and can complete your project. I have expertise in the following: 1- Designing schematic and PCBA 2- Ordering the PCBA and getting boards manufactured. 3- Writing firmware for the PCBA to make sure everything works as expected. 4- I can also develop smartphone apps to communicate and control the hardware. If you wish we can discuss in chat. Thank you.
$41 USD in 7 days
7.4
7.4

I have extensive experience in C programming, Python, electronics, microcontrollers, and debugging. I am confident in my ability to diagnose and resolve the issue with the ArduPilot ExternalNav integration. With access to the C++ firmware on the UWB microcontroller and the current ArduPilot parameters, I will efficiently identify the root cause of the failure and provide updated code or configuration files to ensure a reliable flow of data to ArduPilot. I am excited to work on this project and deliver the required deliverables to your satisfaction. See the above links please. Please go through my profile its 15 years old see the work I did over the years. ---> No Win No Fee means that your satisfaction is my utmost priority. <---- Lets discuss the job details. Moreover, I am willing to start the job and perform tasks without even being hired; it is just to show my commitment to this project. Looking forward to hear from you. Regards Shah
$40 USD in 3 days
6.4
6.4

As an AI systems builder, I am no stranger to integrating complex components for production-level success. My proficiency in C++, Python, and MAVLink will ensure your UWB data makes it smoothly to ArduPilot without experiencing degradation. With deep experience in embedded systems like STM32 -- which powers your UWB microcontroller -- and Raspberry Pi, I know the intricacies of both platforms and can find any bugs within. My track record in firmware development makes me an ideal fit for this project. I've worked extensively on STM32-class platforms and written many C++ firmware codes -- precisely the requirement for this C++ firmware issue you're facing. Being adept with Mission Planner, I can make quick work of debugging and fine-tuning the ArduPilot parameters. Ultimately, my goal is your 100% satisfaction. After all, I'm not just about building software but providing holistic solutions that work reliably - which is exactly what you need for this project. Rest assured that with me on board, you'll have a detailed root cause analysis report in addition to updated code or configuration files that will correct the issue and a verification log or short flight test demonstrating stable VISION_POSITION_ESTIMATE data accepted by ArduPilot
$41 USD in 7 days
6.4
6.4

Hi, I am very interested in helping resolve your ArduPilot ExternalNav integration issue. I have experience with embedded C/C++, Raspberry Pi, MAVLink communication, sensor fusion, and autonomous navigation systems, including debugging data pipelines between external positioning systems and flight controllers. I will systematically trace the complete data path—from the Decawave UWB output through the STM32 firmware, Raspberry Pi MAVLink packaging, and into ArduPilot's EKF—to identify exactly where the position estimates become invalid. I will analyze Mission Planner logs, verify the VISION_POSITION_ESTIMATE messages, review timestamping, coordinate frames, covariance, and EKF parameters, then implement the required code or configuration changes to ensure ArduPilot consistently accepts the ExternalNav data. You will receive a concise root-cause analysis, updated source code or configuration files, and verification logs demonstrating stable ExternalNav operation. I maintain clear communication throughout the project and will work with you until the integration is functioning reliably. I look forward to reviewing your logs and helping you restore a stable ExternalNav pipeline. Regards,
$41 USD in 7 days
5.8
5.8

As an experienced Senior Electronics Engineer, I bring a wealth of skills and knowledge to successfully fix your ArduPilot ExternalNav Integration. With 8 years in the industry, I am well-practiced in resolving complex issues with embedded systems - a skill that will be crucial to pinpointing the fault in your UWB pipeline. My familiarity with C++ in an embedded STM32-class environment, combined with my circuit design and firmware development skills, ensure that I can tackle problems at their root cause. Your project aligns perfectly with what I've accomplished throughout my career: designing and delivering robust, reliable hardware solutions. My experience using Python on Raspberry Pi 5 will assist me in troubleshooting problems specific to your setup; my work will be efficient and effective. Additionally, my track record of working under tight deadlines with multidisciplinary teams speaks for itself - you can expect high-quality results from me and within the specified timeframe. ARDUCEO rights
$140 USD in 5 days
4.4
4.4

Hi, I can debug your ArduPilot ExternalNav pipeline and help identify why the UWB position data is degrading or not being accepted by the EKF. The best solution is to review the full data path first: UWB microcontroller output, Raspberry Pi Python parsing, MAVLink `VISION_POSITION_ESTIMATE` packing, timestamps, coordinate frames, covariance values, message rate, ArduPilot EKF parameters, and Mission Planner logs. I’ll then isolate whether the issue is coming from firmware data, Pi-side conversion, MAVLink formatting, timing, or ArduPilot configuration. I’m comfortable with C/C++ embedded firmware, Python on Raspberry Pi, MAVLink 2.0, ArduPilot 4.x, EKF ExternalNav setup, Mission Planner log review, UWB positioning systems, and bench-test debugging. Deliverables will include: * Root-cause debugging report * Review of current ArduPilot parameters * MAVLink message validation * Timestamp/frame/rate checks * Updated Python/C++ code if needed * Corrected EKF/ExternalNav configuration * Bench-test verification steps * Mission Planner log review * Final verification log showing stable accepted estimates I’ll focus on making the data flow clean, correctly formatted, and reliable so ArduPilot can accept the `VISION_POSITION_ESTIMATE` stream without rejection or instability. Best regards Ankit
$40 USD in 1 day
4.0
4.0

Hi, your ExternalNav chain is failing somewhere between the UWB source, Raspberry Pi 5 MAVLink packing, and ArduPilot’s EKF intake. I’d trace the data end to end, confirm coordinate frames, timestamps, covariance, and message rates, then compare raw telemetry with Mission Planner logs to find where the estimates start degrading. I’ve worked on embedded C++ firmware, Python telemetry pipelines, and MAVLink integration, so I can inspect both the STM32-side code and the Pi bridge without guesswork. I’ll reproduce the bench setup, isolate the fault, then update the code or parameters so ArduPilot receives stable VISION_POSITION_ESTIMATE data it will actually accept. Once the root cause is fixed, I’ll provide a concise report and verification evidence showing the corrected flow in action. Best regards, Gabriel
$40 USD in 5 days
3.6
3.6

The ExternalNav pipeline on ArduPilot can be tricky when MAVLink positioning messages are not landing right in the EKF. I would start by checking the VISION_POSITION_ESTIMATE message timing and confidence values on the Raspberry Pi 5 side, then trace whether EKF2_AID_MASK is set correctly for UWB input. I can start today and have a fix or clear diagnosis within 24 hours. The bid reflects what's in the description. Want to jump on a quick call?
$42 USD in 3 days
3.6
3.6

Hi there, I am a robotics engineer with extensive experience in ArduPilot EKF integration and MAVLink protocol implementation. I understand the frustration of intermittent VISION_POSITION_ESTIMATE rejection—this is typically an EKF variance or synchronization issue, often caused by frame-of-reference mismatches or timestamp jitter between the Pi 5 and the flight controller. My Technical Approach: I will perform a systematic audit of your pipeline to isolate the failure: Data Integrity Check: I will verify the C++ firmware output on your STM32-class UWB hardware to ensure proper packet formatting and consistent frequency before transmission to the Raspberry Pi. Latency & Synchronization: I will audit your Python pipeline on the Pi to check for buffer bloat or latency spikes that may be pushing the EKF "Innovations" outside of the acceptable gating threshold. EKF Tuning: I will review your ArduPilot parameters (VISO_* and EK3_SRC*) to ensure the EKF is correctly configured to weight your UWB data against internal IMU dead reckoning, preventing the "rejection" behavior. Deliverables: You will receive a clear root-cause report, optimized code for the Pi/STM32, and verified logs showing the EKF successfully locking onto your UWB source. Are you seeing specific "EKF Yaw/Position" warning flags in Mission Planner logs? Best regards,
$50 USD in 7 days
5.1
5.1

Hello, I hope this message finds you well. I understand that you're facing challenges with the ExternalNav pipeline, specifically in ensuring accurate position estimates from UWB data to ArduPilot. It’s crucial to resolve these issues for reliable vehicle operation. With extensive experience in embedded systems and real-time data processing, I am well-equipped to tackle this project. My proficiency in C++ for STM32 environments, along with strong skills in Python on Raspberry Pi, positions me to effectively debug your setup. I have worked on similar MAVLink implementations and have a deep understanding of ArduPilot's EKF and ExternalNav configurations. To address the issues you're encountering, I propose the following approach: - Analyze the current pipeline and settings using Mission Planner logs for insights. - Review and test the C++ firmware on the UWB microcontroller to identify potential data degradation points. - Conduct real-time telemetry checks to ensure the integrity of the MAVLink messages. - Deliver a comprehensive report detailing my findings, along with updated code or configuration files that rectify the identified issues. I am eager to help resolve these challenges and ensure a smooth data flow to ArduPilot. I am available to start immediately and would love to discuss any further details to get started. Best regards.
$40 USD in 7 days
3.2
3.2

I WOULD APPRECIATE A MOMENT OF YOUR TIEM. Hi — Abror-Yakubov here from Uzbekistan in Central Asia. A problem well understood is a problem half solved. I have experience with C Programming, Python, C++ Programming and Electronics to debug data pipelines, analyze communication issues, and improve reliable software flows between connected systems. There are three key steps to solve a problem Aren't there? 1. solution1 : Review the ExternalNav pipeline and identify where the position data becomes unreliable. 2. solution2 : Analyze MAVLink messages, firmware logic, and configuration flow to correct the integration issue. 3. solution3 : Verify the updated solution with testing results and provide clear documentation of the fix. Would that be ok with you? And of course don't like your thought. I think that you want clear solution, Quality results for this project and so I will work on these rule for your project, this issue will be successful. May I ask you a question that will help you with your plan? 1. Are the MAVLink VISION_POSITION_ESTIMATE messages currently visible in Mission Planner logs before ArduPilot rejects them? 2. Which part do you suspect most: Raspberry Pi processing, microcontroller firmware, or ArduPilot EKF configuration? Let's discuss in more detail now. I suggest Chat for you
$50 USD in 7 days
3.0
3.0

Hi. thanks for posting this job. I am excited to help I understand your details and know what needs to be done. I will keep the plan simple, ask for feedback. and focus on results. I’m offering best quality and highest performance at lowest price I am an expert with 10 years of experience in C Programming, Python reach their goals. . I have helped many clients bring ideas to life and reach their goals. You can view my portfolio at Hello. I am available now to assist you. Warm regards, bhargav922002
$40 USD in 3 days
0.7
0.7

Hello, does the issue arise immediately upon MAVLink injection or after some time? Understanding this helps diagnose the VISION_POSITION_ESTIMATE degradation. My approach would involve systematically tracing the data flow from the Raspberry Pi to ArduPilot, leveraging your provided logs and telemetry. I've optimized complex systems, like reducing page load times from 3.4s to 2.2s, and can deliver a stable data stream.
$40 USD in 7 days
0.0
0.0

Hi, This is exactly the kind of systems-level debugging work I enjoy. I have experience working with embedded C++, Python, telemetry pipelines, MAVLink integrations, and real-time data systems where the challenge is often identifying where valid sensor data becomes unusable by the flight controller. For this project, I would systematically trace the complete chain: • UWB tag/anchor output validation • STM32 firmware data generation and timestamp handling • Raspberry Pi MAVLink message creation and transport • VISION_POSITION_ESTIMATE formatting and covariance checks • ArduPilot EKF3 and ExternalNav parameter verification • Mission Planner log analysis (EKF innovations, GPS/ExternalNav fusion status, message rates, timing, and frame alignment) Relevant experience includes: • Python-based telemetry and automation systems • Real-time data streaming applications • Robotics and sensor integrations • API and protocol debugging • Performance-critical systems with strict timing requirements My goal would be to identify the exact failure point, provide corrected code/configuration, and validate stable acceptance of VISION_POSITION_ESTIMATE by ArduPilot with supporting logs and verification results. I can begin immediately once access to the firmware, Raspberry Pi code, parameters, and Mission Planner logs is provided.
$41 USD in 2 days
0.0
0.0

Hi, I'm interested in your project and can complete it within a day using MAVLink 2.0, ArduPilot EKF3/ExternalNav, and embedded C++ on STM32. Debugging degraded VISION_POSITION_ESTIMATE feeds from UWB-to-Pi-to-ArduPilot pipelines is exactly the kind of fault-finding I've handled before, usually tracing it to timestamp/covariance mismatches or frame misalignment rather than the UWB data itself. My approach: first pull Mission Planner logs and check EKF3 source selection, VISO_ message timing, and covariance fields against the raw UWB tag output on the Pi. I'll instrument the Python MAVLink packer to log pre-send values, compare against Decawave anchor solve quality, and check for NED/ENU frame or yaw-offset errors that silently corrupt position fusion. Likely culprits are stale timestamps causing EKF rejection or bad covariance scaling making ArduPilot distrust the estimate — I'll isolate which with a bench replay before touching firmware. One thing worth flagging: don't just increase EKF innovation gates to force acceptance — that masks real anchor geometry or multipath issues instead of fixing them. Can you confirm the UWB tag update rate and whether timestamps are synced to the Pi's clock or free-running? Ready to start as soon as you share the logs. Best, Emrah
$41 USD in 1 day
0.0
0.0

Hi! I work with Python on Raspberry Pi and MAVLink-based systems regularly, including real-time sensor data pipelines and microcontroller integration. Debugging VISION_POSITION_ESTIMATE flow into ArduPilot EKF is something I can tackle systematically from the logs and telemetry you have ready. Share the Mission Planner logs, Python script, and ArduPilot parameters and I can start immediately.
$42 USD in 2 days
0.0
0.0

Hi There, This looks like an interesting debugging task, and I'd be happy to help. I have experience working with C++, Python, MAVLink, and embedded systems, and I understand how sensitive the ExternalNav pipeline can be when passing data into ArduPilot's EKF. I'll trace the complete data flow—from the UWB firmware, through the Raspberry Pi MAVLink bridge, all the way to ArduPilot—to identify exactly where the position estimates are breaking down. Once I have access to your firmware, parameters, Mission Planner logs, and bench setup, I'll: * Inspect the MAVLink `VISION_POSITION_ESTIMATE` messages for timing, coordinate frames, timestamps, and data integrity. * Review the Raspberry Pi bridge and ArduPilot EKF/ExternalNav configuration. * Identify the root cause, implement the required code or parameter fixes, and verify that ArduPilot consistently accepts the position estimates. You'll receive a clear explanation of the issue, the updated code/configuration, and verification logs (or a bench/flight test) confirming the fix. Based on our final discussion, I'll provide the timeline and a fixed cost for the project before we get started. Regards, Palak
$41 USD in 3 days
0.0
0.0

Hi, I have hands on experience developing UAV communication and embedded systems, including integrating sensor data pipelines, embedded firmware, Raspberry Pi-based processing, and real-time communication interfaces. For your UWB ExternalNav issue, I can systematically debug the complete chain from STM32 UWB firmware and data quality, through Raspberry Pi MAVLink packaging, to ArduPilot EKF3 acceptance. I will analyze logs, verify coordinate frames, timestamps, covariance, update rates, and EKF parameters to identify the exact failure point. My approach is to build a reliable ExternalNav pipeline with validated MAVLink VISION_POSITION_ESTIMATE output, proper filtering, and stable EKF integration rather than only applying temporary parameter changes. I can provide: - Root cause analysis report - Corrected firmware, software, configuration - Verified MAVLink and ArduPilot integration results I am ready to review your code, parameters, and logs and start isolating the issue immediately. Best regards.
$40 USD in 5 days
0.0
0.0

Hello, I have worked on UAV electronics, embedded firmware, wireless communication systems, and real-time sensor integration, with experience bridging hardware data into reliable control systems. Your UWB ExternalNav problem is exactly the type of integration challenge I handle tracing the complete data path from the UWB tag and STM32 firmware, through Raspberry Pi MAVLink processing, into ArduPilot EKF3. I will verify the data integrity, timing, coordinate transformations, covariance handling, and EKF configuration to find the real cause of rejection. My goal is not only to make the system pass a test, but to deliver a clean and repeatable ExternalNav pipeline that ArduPilot can trust during operation. I can provide a clear root-cause report, corrected code/configuration, and validation results from bench or flight testing. Looking forward to helping you stabilize the system. Best regards.
$50 USD in 2 days
0.0
0.0

Kuala Lumpur, Malaysia
Payment method verified
Member since Nov 9, 2025
$30-250 USD
$10-50 USD
₹1500-12500 INR
$250-750 USD
$10-30 USD
$250-750 USD
$30-250 CAD
₹12500-37500 INR
$30-250 USD
₹600-1500 INR
₹1500-12500 INR
$30-250 CAD
$30-250 USD
₹1500-12500 INR
$8-15 USD / hour
₹12500-37500 INR
₹600-1500 INR
$30-250 USD
$30-250 USD
$2-8 USD / hour
₹12500-37500 INR
$10-30 USD