DrunkWalk

We present LiLoc , a system for precise 3D localization and tracking of mobile IoT devices (e.g., robots) in indoor environments using multi-perspective LiDAR sensing. LiLoc stands out with two key differentiators. First, unlike traditional localization approaches, our method remains robust in dynamically changing environments, adeptly handling varying crowd levels and object layout changes. Second, LiLoc is independent of pre-built static maps, employing dynamically updated point clouds from infrastructural-mounted LiDARs and LiDARs on individual IoT devices. For fine-grained, near real-time tracking, LiLoc intermittently utilizes complex 3D “global” registration between point clouds for robust spot location estimates. It further complements this with simpler “local” registrations, continuously updating IoT device trajectories. We demonstrate that LiLoc can (a) support accurate location tracking with location and pose estimation error being ≦7.4 cm and ≦3.2°, respectively, for 84% of the time and the median error increasing only marginally (8%), for correctly estimated trajectories, when the ambient environment is dynamic; (b) achieve a 36% reduction in median location estimation error compared to an approach that uses only quasi-static global point cloud; and (c) obtain spot location estimates with a latency of only 973 msec. We also demonstrate how LiLoc efficiently integrates low-power inertial sensing, using a novel integration of inertial-based displacement to accelerate the local registration process, to enhance localization energy efficiency and latency.

We present a quadrotor Micro Aerial Vehicle (MAV) capable of autonomous indoor navigation. The MAV is equipped with four cameras arranged in two stereo configurations. One camera pair is facing forward and serves as input for a reduced stereo SLAM system. The other camera pair is facing downwards and is used for ground plane detection and tracking. All processing, including sparse stereo matching, is run on-board in real-time and at high processing rates. We demonstrate the capabilities of this MAV design in several flight experiments. Our MAV is able to recover from pose estimation errors and can cope with processing failures for one camera pair. We show that by using two camera pairs instead of one, we are able to significantly increase navigation accuracy and robustness.

This paper addresses the problem of autonomous navigation of a micro air vehicle (MAV) in GPS‐denied environments. We present experimental validation and analysis for our system that enables a quadrotor helicopter, equipped with a laser range finder sensor, to autonomously explore and map unstructured and unknown environments. The key challenge for enabling GPS‐denied flight of a MAV is that the system must be able to estimate its position and velocity by sensing unknown environmental structure with sufficient accuracy and low enough latency to stably control the vehicle. Our solution overcomes this challenge in the face of MAV payload limitations imposed on sensing, computational, and communication resources. We first analyze the requirements to achieve fully autonomous quadrotor helicopter flight in GPS‐denied areas, highlighting the differences between ground and air robots that make it difficult to use algorithms developed for ground robots. We report on experiments that validate our solutions to key challenges, namely a multilevel sensing and control hierarchy that incorporates a high‐speed laser scan‐matching algorithm, data fusion filter, high‐level simultaneous localization and mapping, and a goal‐directed exploration module. These experiments illustrate the quadrotor helicopter's ability to accurately and autonomously navigate in a number of large‐scale unknown environments, both indoors and in the urban canyon. The system was further validated in the field by our winning entry in the 2009 International Aerial Robotics Competition, which required the quadrotor to autonomously enter a hazardous unknown environment through a window, explore the indoor structure without GPS, and search for a visual target. © 2011 Wiley Periodicals, Inc.

Performing autonomous navigation of commercial Micro Air Vehicles (MAVs) in GPS-denied environments without external sensors or a motion capture system is a major challenge in trajectory tracking applications. A convenient solution for this problem is the use of the MAV’s odometry, which uses inertial sensors for pose estimation. However, it can suffer from position drifting over time, which may lead to incorrect pose measurements. Another solution is to use the MAV’s monocular camera for pose estimation, based on visual Simultaneous Localization and Mapping (vSLAM) algorithms. A limitation of this approach is that monocular vSLAM lacks a metric scale, hence, making its use unfeasible for position feedback. In this scenario, we propose a robust navigation system of a MAV based on vSLAM where a metric scale can be estimated using a Kalman Filter (KF), based on odometry information. In this work, we opted for the use of a Robust Linear-Quadratic Regulator (R-LQR), which is a recursive strategy that takes into account parametric uncertainties on the system’s dynamic. We present simulated and experimental results of our approach using a commercial quadrotor to show the effectiveness of the system.

With the rapid development of UAV industry, autonomous UAV obstacle avoidance navigation has become a core problem in the field of UAV control. Based on recurrent neural networks, this paper proposes an LSTM-enhanced Layered-RSAC algorithm to construct a collaborative task execution and autonomous navigation system for UAV swarms. By constructing the autonomous navigation system of UAV, its accuracy is tested and the UAV operation situation index is examined. Through model training, the standard deviation of the a priori strategy with the best success rate of autonomous navigation is explored. The a priori strategy σ = 0.45 is taken as the initial value to verify the performance improvement of the Layered-RSAC algorithm. The results show that the Layered-RSAC algorithm reaches 90% navigation success rate at 50 training steps for the first time and stabilizes at 90% to 100% success rate after 100 training steps, which is significantly ahead of Prior-Policy, DDPG and SAC algorithms.

Autonomous navigation for large Unmanned Aerial Vehicles (UAVs) is fairly straight-forward, as expensive sensors and monitoring devices can be employed. In contrast, obstacle avoidance remains a challenging task for Micro Aerial Vehicles (MAVs) which operate at low altitude in cluttered environments. Unlike large vehicles, MAVs can only carry very light sensors, such as cameras, making autonomous navigation through obstacles much more challenging. In this paper, we describe a system that navigates a small quadrotor helicopter autonomously at low altitude through natural forest environments. Using only a single cheap camera to perceive the environment, we are able to maintain a constant velocity of up to 1.5m/s. Given a small set of human pilot demonstrations, we use recent state-of-the-art imitation learning techniques to train a controller that can avoid trees by adapting the MAVs heading. We demonstrate the performance of our system in a more controlled environment indoors, and in real natural forest environments outdoors.

Millimeter wave (mmWave) radars have attracted significant attention from both academia and industry due to their capability to operate in extreme weather conditions. However, they face challenges in terms of sparsity and noise interference, which hinder their application in the field of micro aerial vehicle (MAV) autonomous navigation. To this end, this letter proposes a novel approach to dense and accurate mmWave radar point cloud construction via cross-modal learning. Specifically, we introduce diffusion models, which possess state-of-the-art performance in generative modeling, to predict LiDAR-like point clouds from paired raw radar data. We also incorporate the most recent diffusion model inference accelerating techniques to ensure that the proposed method can be implemented on MAVs. We validate the proposed method through extensive benchmark comparisons and real-world experiments, demonstrating its superior performance and generalization ability.

This letter proposes a novel Nonlinear Model Predictive Control (NMPC) framework for Micro Aerial Vehicle (MAV) autonomous navigation in indoor enclosed environments. The introduced framework allows us to consider the nonlinear dynamics of MAVs, nonlinear geometric constraints, while guarantees real-time performance. Our first contribution is to reveal underlying planes within a 3D point cloud, obtained from a 3D lidar scanner, by designing an efficient subspace clustering method. The second contribution is to incorporate the extracted information into the nonlinear constraints of NMPC for avoiding collisions. Our third contribution focuses on making the controller robust by considering the uncertainty of localization in NMPC using Shannon's entropy to define the weights involved in the optimization process. This strategy enables us to track position or velocity references or none in the event of losing track of position or velocity estimations. As a result, the collision avoidance constraints are defined in the local coordinates of the MAV and it remains active and guarantees collision avoidance, despite localization uncertainties, e.g., position estimation drifts. The efficacy of the suggested framework has been evaluated using various simulations in the Gazebo environment.

Typical industrial applications, which can benefit from using Micro Aerial Vehicles (MAV) take place at well-controlled environments. In this case, the MAVs can rely on a fiducial system to improve their navigation based on a reliable map of the environment. In this work, we assess the use of artificial landmarks for accurate and robust visual-inertial navigation within a controlled environment. The use of artificial landmarks presents relevant advantages over techniques that assume an unknown environment. They require less processing effort and provide more robustness for dynamic and large environments. As a result, one can employ simpler, affordable platforms enabling commercially attractive applications due to cost reduction, energy efficiency, simplicity and determinism during operation. Therefore, this work presents the design and development of an accurate and efficient visual-inertial navigation system to provide feedback information allowing low and high-level control functionalities for MAV applications.

In this paper, indoor environment classification and interpretation algorithm is proposed. Proposed algorithm needs low computation power and low payload thus enabling micro air vehicle (MAV) to quickly react and navigate. Here indoor environment is classified into corridor, staircase, and open space by using image edge gist descriptors and a neural network classifier. Use of some predetermined thresholds further increases the confidence of the classification and interpretation algorithm. Detection of horizontal lines cluster and vanishing point is used for the navigation in staircase and corridor environment respectively. Results demonstrate that the proposed algorithm can interpret the indoor environment effectively with > 90% accuracy.

Essential for achieving fully autonomous igh t and good igh t performances of Mini and Micro Aerial Vehicles (MAV) is a reliable determination of position, velocity and attitude of the aircraft. In the institute its current MAV navigation concept, an IMU providing acceleration and angular rate measurements at 100 Hz is used for this purpose among other systems. An important task of the IMU is the supply of angular velocities so that a high-frequency attitude solution can be computed. A high attitude update frequency is necessary because of the high rotational dynamics of MAVs which can be less than 0.1 s. Due to the small size and weight of a MAV only miniaturized sensors are applicable. Unfortunately, such small and usually silicon-based inertial sensors (micro-electromechanical systems, MEMS) have signican t and strong temperature correlated deterministic errors. To overcome the temperature inuence, temperature was taken into account during sensor calibration. After calibration of the sensor pack a test igh t with the university’s research aircraft, a Dornier Do 128-6, was carried out. The aircraft was equipped with a high-precision INS Honeywell Lasernav and a Novatel OEM3 GPS receiver as reference sensors. The postprocessing of the measurement data shows that the calibration results can be conrmed by the reference sensors. First and second order approximations for the calibration tables turned out to provide equivalent navigation accuracy compared to the original calibration tables while saving computational power. Satellite and inertial navigation data were integrated within a tightly-coupled Kalman Filter. In order to improve the navigation performance especially during loss of GPS the attitude information derived from on-board camera images is used to aid the lter. The results presented are based on real igh t measurement data for the MEMS-based sensors and on simulated GPS measurements.

In this study, we investigate the fundamental limits for time delay based location estimation in decode-andforward (DF) relaying strategy based cooperative relay networks considering multipath environments. The limits are presented by obtaining Cramer-Rao Lower Bounds (CRLBs) expressions. Simulations are performed to verify the CRLB expressions and show the effects of the node topology of the network, number of relay nodes, and number of multipath components (MPCs) on the location estimation accuracy of source node. According to the results, the placement of the relay nodes is very important to lower the location estimation error. Furthermore, using more relay nodes decreases the location estimation error, dramatically. Finally, processing more MPCs at the destination node is crucial to improve the estimation accuracy in the cooperative relay networks.

We present an open‐source system for Micro‐Aerial Vehicle (MAV) autonomous navigation from vision‐based sensing. Our system focuses on dense mapping, safe local planning, and global trajectory generation, especially when using narrow field‐of‐view sensors in very cluttered environments. In addition, details about other necessary parts of the system and special considerations for applications in real‐world scenarios are presented. We focus our experiments on evaluating global planning, path smoothing, and local planning methods on real maps made on MAVs in realistic search‐and‐rescue and industrial inspection scenarios. We also perform thousands of simulations in cluttered synthetic environments, and finally validate the complete system in real‐world experiments.

Location estimation in wireless sensor networks (WSNs) has received tremendous attention in recent times. Improved technology and efficient algorithms systematically empower WSNs with precise location identification. However, while algorithms are efficient in improving the location estimation error, the factor of the network lifetime has not been researched thoroughly. In addition, algorithms are not optimized in balancing the load among nodes, which reduces the overall network lifetime. In this paper, we have proposed an algorithm that balances the load of computation for location estimation among the anchor nodes. We have used vector-based swarm optimization on the connected dominating set (CDS), consisting of anchor nodes for that purpose. In this algorithm, major tasks are performed by the base station with a minimum number of messages exchanged by anchor nodes and unknown nodes. The simulation results showed that the proposed algorithm significantly improves the network lifetime and reduces the location estimation error. Furthermore, the proposed optimized CDS is capable of providing a global optimum solution with a minimum number of iterations.

Localization of the nodes in wireless networks is crucial for supporting the location-based services (LBSs). In this study, we investigate the fundamental limits for time delay based location estimation in amplify-and-forward (AF) cooperative relay networks considering multipath environment. The limits are presented by obtaining Cramer-Rao Lower Bounds (CRLBs) expressions. Simulations are performed to verify the CRLB expressions and show the effects of node topology of the network, number of relay nodes, and number of multipath components (MPCs) on the location estimation of source node. According to the results, the placement of the relay nodes is very important to lower the location estimation error. Furthermore, using more relay nodes decreases the location estimation error, dramatically. Finally, processing more MPCs at the destination node is crucial to improve the estimation accuracy for both relaying strategy.