inertial navigation system
PulseAugur coverage of inertial navigation system — every cluster mentioning inertial navigation system across labs, papers, and developer communities, ranked by signal.
10 day(s) with sentiment data
IMU data is increasingly being fused with other modalities for complex tasks
Recent research highlights the integration of IMU data with egocentric video for motor impairment assessment and with EMG for advanced gesture recognition. This trend suggests a growing recognition of IMU's value as a complementary sensor, moving beyond standalone applications towards richer, multimodal sensing solutions.
Quantum-enhanced INS to see commercialization within 1-2 years
Lianggan Intelligence, a startup focused on quantum sensing for inertial navigation, has secured significant angel funding. Given their proprietary technology and early application in INS, it's plausible they will bring a commercial product to market within the next 1-2 years, potentially disrupting the high-precision navigation sector.
Improved IMU data processing for exercise evaluation to reach commercial viability
A new method addresses label ambiguity in IMU-based exercise evaluation by generating label distributions instead of single labels. This advancement, which matches or exceeds baseline performance, indicates a path towards more robust and accurate commercial fitness tracking and rehabilitation tools.
-
EventHOI system enables efficient hand-object interaction logging on smartglasses
Researchers have developed EventHOI, a novel system designed to efficiently log hand-object interactions (HOIs) from the perspective of smartglasses. This system addresses the limitations of continuous RGB video capture…
-
FREE-LIVO method decouples LiDAR-camera sensors for robust odometry
Researchers have developed FREE-LIVO, a novel LiDAR-Inertial-Visual Odometry method that decouples the rigid mounting and precise time synchronization typically required between LiDAR and camera sensors. This approach u…
-
New MotorSense dataset links video and EMG for enhanced action understanding
Researchers have introduced MotorSense, a novel dataset designed to enhance action understanding in computer vision models by incorporating underlying motor signals. This dataset pairs egocentric video with synchronized…
-
IMUdrape generates virtual IMU data for garment sensors using physics simulation
Researchers have developed IMUdrape, a novel system capable of generating virtual Inertial Measurement Unit (IMU) data from text descriptions for garment-mounted sensors. Unlike previous methods that rigidly attach virt…
-
New dataset enables ECG reconstruction from vibrational signals
Researchers have developed Vib2ECG, a new dataset and benchmark for reconstructing electrocardiography (ECG) signals from vibrational data. This approach aims to enable more accessible, mobile-device-friendly ECG monito…
-
Transformer-based TLEIO achieves precise motion from event cameras
Researchers have developed Tight Learned Event-Inertial Odometry (TLEIO), a novel Transformer-based architecture designed for precise motion estimation from event camera data. This system, named EventsFormer, processes …
-
Insta360 cameras' embedded IMUs offer accurate metric scale for SfM
A new paper proposes using the embedded Inertial Measurement Unit (IMU) in Insta360 X5 and X6 cameras to achieve metric scale for structure-from-motion (SfM) tasks. This method offers an alternative to relying on smartp…
-
Machine learning applied directly to sensor data for embedded systems
This episode of Embedded.fm, titled "I've Seen A Giraffe," features David discussing the limitations of raw sensor data, such as Inertial Measurement Units (IMUs). He explains that this raw data is insufficient on its o…
-
Electronics-free glove uses single head camera for robot control
Researchers have developed an electronics-free glove that can be labeled and observed using a single head-mounted camera, eliminating the need for separate pose trackers and multiple data streams. This approach simplifi…
-
OmniPoser framework reconstructs 3D human motion from flexible sensor inputs
Researchers have introduced OmniPoser, a novel framework designed to reconstruct 3D human motion from a flexible array of sensor inputs. Unlike previous methods that require specific sensor configurations and are comput…
-
FALLNEST framework condenses IMU data for wearable fall detection
Researchers have developed FALLNEST, a novel data-centric framework for wearable fall detection using Inertial Measurement Unit (IMU) data. This approach addresses the challenges of imbalanced datasets and the computati…
-
MultiFly dataset offers multimodal aerial perception benchmarks
Researchers have introduced MultiFly, a novel real-world dataset designed for multimodal aerial perception using unmanned aerial vehicles (UAVs). The dataset features synchronized data across RGB, thermal, LiDAR, and ra…
-
New CARAT method improves multimodal time series adaptation
Researchers have developed CARAT, a novel method for enhancing the reliability of multimodal time series data, particularly in wearable systems. CARAT decouples model reliance from runtime corruption detection, using a …
-
New AI model reconstructs ECG signals from wearable sensors
Researchers have developed BeatFlow-ECG, a novel model capable of reconstructing electrocardiography (ECG) signals from less invasive wearable sensors like photoplethysmogram (PPG) and inertial measurement units (IMU). …
-
Open-source ALFRED robot enhances plant monitoring with improved reach
Researchers have developed ALFRED, an open-source mobile manipulator designed for long-term plant monitoring in outdoor environments. Built from readily available components, ALFRED features a six-degree-of-freedom arm,…
-
Machine learning detects e-scooter rider intoxication via kinematic analysis
Researchers have developed a method to detect alcohol intoxication in e-scooter riders using sensor data and machine learning. By instrumenting an e-scooter with an Inertial Measurement Unit (IMU) and other sensors, the…
-
New SPAR system analyzes boxing technique with wearable sensors
Researchers have developed a new system called Skill Profiling with Attributable Reasoning (SPAR) designed to analyze boxing techniques using wearable sensors. This system, comprising an eight-IMU garment and pressure i…
-
Single earbud IMU can recover full-body 3D pose, study finds
Researchers have developed a method to estimate full-body 3D pose using data from a single consumer earbud's Inertial Measurement Unit (IMU). Experiments showed that a single earbud IMU could recover lower-body pose wit…
-
New framework enhances human activity recognition with virtual IMU data
Researchers have developed a new framework for improving human activity recognition (HAR) using wearable Inertial Measurement Units (IMUs). The proposed method addresses the challenge of limited labeled IMU data by inte…
-
LiDAR-only cone detection framework runs on CPU for driverless racing
Researchers have developed a lightweight, LiDAR-only perception system for Formula Student Driverless vehicles that runs efficiently on a CPU. This system utilizes a Random Forest classifier, ground removal, IMU-based m…