Fig. 2 General form of a feedback complementary Kalman filter. A specific example of this approach for the multivehicle formulation can be found in Fig. 4 . More

Fig. 1 Multimodal reconnaissance and security operations of joint manned and unmanned systems [ 23 ] More

Fig. 3 Overview of team architecture. The ground vehicles observe the UAV in flight using their stereoscopic camera systems. The UAV observes objects in the environment using a downward-facing camera. Communication between vehicles and a ground station PC occurs through a local WiFi network. The g... More

Fig. 4 Indirect CKF method for collaborative localization of a ground vehicle. This method fuses the measurements of the five independent signals arranged in the column on the left. The signals s 1 , s 3 , and s 5 represent the estimated pose of each vehicle combining the most recent state ... More

Fig. 5 Circle detection of spherical markers for triangulation: ( a ) original, ( b ) U space, ( c ) V space, ( d ) boolean, ( e ) circled, and ( f ) colored, spherical markers and corresponding axes of the UAV defined by the markers. More

Fig. 6 Simulation and experiment trajectories of UAV controlled by static UGV. Thick line segments represent he goal poses for the UAV and the thin solid lines are the ground-truth measurements from the Vicon motion capture system. The bottom dashed lines are the estimated trajectory of the UAV in... More

Fig. 7 A still frame from experiment simulation at the moment that a plan for the UGV has been computed but before the UGV begins to move More

Fig. 8 (Top) UGV estimated and actual trajectory for trial 5. Coordinates are relative to the UGV’s starting orientation. The dashed circles represent the circumscribed physical footprint of the UGV and the goal landmark. (Bottom) Error of UGV estimate as a function of distance traveled. More

Fig. 9 Trajectories of UGV for trials 1–5. The dashed circles circumscribe the physical footprint of the robot and the goal landmark. Note: obstacle colors correspond to the color of the trial in which they are present. More

Fig. 10 Problem setup: the team of two UGVs and one UAV after a goal landmark has been identified and UGV1 has planned a path to the goal More

Fig. 11 Combined RMS errors of the UAV and both UGV as a function of time on ( a ) and ( b ) map 1, and on ( c ) and ( d ) map 4. Errors are averaged across all trials for given strategies and map. ( a ) Strategy 1 C H : UAV hovers over goal location, map 1. ( b ) Strategy 1 C L : UAV leads ... More

Fig. 12 RMS error of UAV and UGV as functions of time (strategy 2CLR, map 4). Errors are averaged across all trials for a given strategy and map. Circle detection of spherical markers for triangulation. ( a ) CKF results for strategy 2 C R L . ( b ) SLAM results for strategy 2 ... More

Fig. 1 Body and inertial frames of a quadcopter More

Fig. 2 Roll maneuver: Euler angles comparison with roll, pitch, and yaw as ϕ , θ , ψ , respectively More

Fig. 3 Roll maneuver: motor spin rates More

Fig. 4 Integration of state equations: ( a ) Position, ( b ) Velocity, ( c ) Quaternions, and ( d ) Angular velocity More