Addressing Stability Challenges in High-Speed Drones

Researchers at Hong Kong University MaRS Lab have just published another jaw dropping paper featuring their safety-assured high-speed aerial robot path planning system dubbed "SUPER". With a single MID360 lidar sensor they repeatedly achieved autonomous one-shot navigation at speeds exceeding 20m/s in obstacle rich environments. Since it only requires a single lidar these vehicles can be built with a small footprint and navigate completely independent of light, GPS and radio link. This is not just #SLAM on a #drone, in fact the SUPER system continuously computes two trajectories in each re-planning cycle—a high-speed exploratory trajectory and a conservative backup trajectory. The exploratory trajectory is designed to maximize speed by considering both known free spaces and unknown areas, allowing the drone to fly aggressively and efficiently toward its goal. In contrast, the backup trajectory is entirely confined within the known free spaces identified by the point-cloud map, ensuring that if unforeseen obstacles are encountered or if the system’s perception becomes uncertain, the system can safely switch to a precomputed, collision-free path. The direct use of LIDAR point clouds for mapping eliminates the need for time-consuming occupancy grid updates and complex data fusion algorithms. Combined with an efficient dual-trajectory planning framework, this leads to significant reductions in computation time—often an order of magnitude faster than comparable SLAM-based systems—allowing the MAV to operate at higher speeds without sacrificing safety. This two-pronged planning strategy is particularly innovative because it directly addresses the classic speed-safety trade-off in autonomous navigation. By planning an exploratory trajectory that pushes the speed envelope and a backup trajectory that guarantees safety, SUPER can achieve high-speed flight (demonstrated speeds exceeding 20 meters per second) without compromising on collision avoidance. If you've been tracking the progress of autonomy in aerial robotics and matching it to the winning strategies emerging in Ukraine, it's clear we're likely to experience another ChatGPT moment in this domain, very soon. #LiDAR scanners will continue to get smaller and cheaper, solid state VSCEL based sensors are rapidly improving and it is conceivable that vehicles with this capability can be built and deployed with a bill of materials below $1000. Link to the paper in the comments below.

𝗗𝗿𝗼𝗻𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁 𝗶𝘀 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝗳𝗹𝗶𝗴𝗵𝘁 — 𝗶𝘁’𝘀 𝗮 𝗳𝘂𝗹𝗹 𝗲𝗺𝗯𝗲𝗱𝗱𝗲𝗱 𝗲𝗰𝗼𝘀𝘆𝘀𝘁𝗲𝗺. Behind every stable flight is a system designed to survive gravity, vibration, packet loss, and sensor noise in real time. 𝗖𝗼𝗿𝗲 𝗘𝗺𝗯𝗲𝗱𝗱𝗲𝗱 𝗕𝗹𝗼𝗰𝗸𝘀 𝗶𝗻 𝗮 𝗗𝗿𝗼𝗻𝗲: 💠Flight Controller (MCU/RTOS-based). 💠Sensor Fusion (IMU, GPS, magnetometer). 💠Motor Control (PWM, ESC, PID loop). 💠Communication Module (RF/LoRa/4G). 💠Failsafe Systems (GPS lock, altitude failback, return-to-home). 💠Power Monitoring (LiPo battery sensing + protection logic). 🔺Challenges in R&D: ✳️Tuning PID in unstable wind. ✳️Syncing ESCs with minimal jitter. ✳️Dealing with brownout resets in mid-air. ✳️Latency in live video + command feedback. ✳️EMI from motors affecting IMU reads. ✳️Integrating AI at the edge. (target lock, tracking, collision avoidance). > “Building a drone isn’t just about flying-it’s about orchestrating dozens of real-time systems to keep flying.” #DroneDevelopment #EmbeddedSystems #RTOS #MotorControl #SensorFusion #FlightController #FirmwareEngineering #EdgeAI #PhDThoughts #LoRa #Quadcopters #PIDTuning #Embeddedc #Embedded #Linux #OS

New FPV Drone Sets 374 MPH Speed Record Drone Pro Hub has pushed drone performance into a new league. Their latest custom FPV machine reached a verified top speed of 374 mph (603.47 km/h), roughly Mach 0.49. That breaks previous unofficial quadcopter speed records and sets a new milestone for FPV engineering. This build now claims the record formerly held by Peregreen 3, the craft developed by Luke Maximo Bell and his father. Peregreen 3 had reached 585 km/h, and its attempt had been widely covered on DroneXL by our friend Zachary Peery just weeks ago. Bell himself described Peregreen as built solely to go “as fast as physically possible.” Photo credit: DroneXL At Drone Pro Hub, the goal was more than setting a number. Their engineers wanted to understand how a drone behaves at 167 meters per second. At those speeds, airflow, vibrations, power systems, and control dynamics all shift dramatically. Motors, ESCs, batteries, the frame and electronics all get tested in ways normal FPV drones never see. The team argues the lessons learned can improve stability and performance even for slower drones. Seventeen Months of Design and Testing The record craft was not based on standard off-the-shelf racing gear. Engineer Ben Biggs and the Drone Pro Hub team designed the entire drone from scratch using CAD models. In the first eight months they developed initial designs, built early prototypes, and ran basic tests reaching 200–300 km/h. At this stage they learned about balance, airflow, structural stress, motor loads, propeller dynamics. Between months nine and twelve they stepped up testing. The drone flew over 30 test flights. Several frames were rebuilt because parts failed. Motors and ESC units overheated under stress. One prototype was destroyed beyond repair. The cost of lost components exceeded $3,000. Then they built a second prototype, stronger and more refined, with improved sensors and better layout. The real breakthrough came between months thirteen and sixteen. Analyzing flight data revealed that the drone’s nose and body contour created too much aerodynamic drag. By redesigning the shape — slimming the nose and smoothing the body — they cut drag by about 18 percent. After that, the drone hit speeds around 540 km/h for the first time. The Record Flight Finally, in the seventeenth month, everything was ready. The drone was stable, all components had passed stress tests, and weather was optimal. The team performed final checks: balanced props, preheated batteries, validated control systems, monitored motor temperature and frame integrity. https://lnkd.in/eVa-M3gK When the drone launched, telemetry showed stable flight under full throttle. Unlike some earlier high-speed attempts (for example the water-cooled ESC approach tried by Peregreen this build handled power and heat without exotic cooling — the improved design and components handled stress natively. The drone accelerated smoothly, maintained stable f...

𝐀𝐒𝐂𝐄 𝐢𝟑𝐂𝐄 𝟐𝟎𝟐𝟓 𝐔𝐩𝐝𝐚𝐭𝐞𝐬 𝟭𝟭 𝗮𝗻𝗱 𝟭𝟮 Two impactful presentations by Tianyu Ren, tackling the critical challenges of UAV stability and precision tool handling in construction operations, pushing the boundaries of drone-assisted automation for the built environment. 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗥𝗼𝗯𝗼𝘁𝗶𝗰 𝗖𝗼𝗻𝘁𝗿𝗼𝗹 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀 𝗳𝗼𝗿 𝗨𝗔𝗩 𝗦𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝘄𝗶𝘁𝗵 𝗥𝗼𝘁𝗮𝘁𝗼𝗿𝘆 𝗣𝗮𝘆𝗹𝗼𝗮𝗱 Tianyu opened the session with a compelling study on a hybrid control framework combining predictive algorithms and machine learning to address the flight instability caused by rotating payloads. By fusing real-time sensor inputs with adaptive control methods, the approach helps UAVs maintain balance and precision during demanding construction tasks. Experimental validation demonstrated improved disturbance rejection and increased payload handling capability, marking a significant step toward practical deployment in the field. 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗚𝗶𝗺𝗯𝗮𝗹 𝗦𝘁𝗮𝗯𝗶𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗧𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲𝘀 𝗳𝗼𝗿 𝗨𝗔𝗩𝘀 𝗶𝗻 𝗛𝗶𝗴𝗵-𝗣𝗿𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝗖𝗼𝗻𝘀𝘁𝗿𝘂𝗰𝘁𝗶𝗼𝗻 𝗧𝗮𝘀𝗸𝘀 The second presentation introduced an adaptive gimbal system engineered to enhance UAV tool control during surface finishing and manipulation tasks. This system actively compensates for environmental factors such as wind, UAV motion, and tool vibrations, achieving superior stability and task accuracy. The findings highlight the potential of gimbal-based solutions to overcome one of the biggest hurdles in precision aerial construction, paving the way for broader adoption of drones in field applications. Congratulations to Tianyu for contributing valuable insights and solutions to the future of aerial robotics in construction. Stay tuned for the full papers in the upcoming ASCE i3CE 2025 Proceedings!