LiDAR is one of the most used sensors in many areas like robotics, self-driving cars, and advanced driving assistance systems due to providing an accurate point cloud of the surroundings. However, to cope with challenges in perceiving the environment around a vehicle, LiDAR data is often combined with data from other sensors. Thermal cameras can provide complementary information that can be beneficial, especially for detecting pedestrians and seeing at nighttime and in fog, dust, etc. In this paper, we propose an algorithm for the extrinsic calibration of a thermal camera and a LiDAR sensor in a vehicle. First, one or more thermal image-point cloud pairs of our designed calibration target are collected. Then line and plane equations of the target’s edges and plane in both data modalities are found. Finally, the algorithm uses lines and plane correspondences to cross-calibrate the sensors. The proposed method obtains good results with one or more poses. We also show that it works well with sparse LiDAR data. Several experiments are presented to illustrate the effectiveness of the method.

Kinematic calibration of a robotic manipulator determines its fabricated dimensions that inevitably deviate, due to manufacturing tolerances, from the specified/designed geometry. Conventional calibration methods achieve this goal by deriving a system of constraint equations from the kinematic structure. Workspace coordinates and the corresponding joint space coordinates are experimentally obtained so that these constraint equations can be solved for the optimal geometric parameters that best fit the data. This paper proposes an alternative kinematic calibration method that leverages response surface methodology to obtain empirical models of direct and inverse kinematics from the same experimental data. The advantages the proposed method include nullifying the requirements of deriving constraint equations and performing joint sensor calibration. In addition, since the developed models are empirical in nature, whether closed-form solutions to the direct or the indirect model is available becomes conveniently inconsequential. Both physical and simulation experiments were conducted on a 3-PSS/S to evaluate the accuracy of the proposed method.

This paper details the development of a tracking controller for a highly dynamic parallel orientation manipulator that is capable of achieving high angular acceleration. The adopted control algorithm is derived from the active disturbance rejection control (ADRC) technology. The ADRC algorithm must be evaluated on the control hardware at a frequency consistent with the dynamics of the system without creating any inter-channel latency for a successful implementation. In this regard, the field programmable gate array (FPGA) is preferred because of its superior speed and parallelism. While there is a clear demand for FPGA controllers in the military and the aerospace domain because of the improved SWaP (size, weight, and power) capabilities, the literature provides a few examples of such controllers that mostly employ expensive high logic density FPGA chips. In contrast, this paper describes how an advanced controller can be efficiently prototyped on a relatively low-cost, low logic density FPGA hardware. The controller was tested in a co-simulation approach (wherein the digital implementation of the control algorithm and the dynamics of the manipulator are simulated together) to ensure a robust FPGA implementation. In addition, a hardware evaluation was also conducted using a linear voice coil actuator under varying inertial loads. Experimental results obtained from the simulation study and hardware testing confirm the robust performance of the designed controller.

Voice-coil actuators are the simplest form of electric motor consisting of a non-commutated single coil or winding moving through a fixed magnetic field produced by stationary permanent magnets. From a system design point of view, however, it is generally the end user's responsibility to couple the voice-coil actuator with a linear bearing system, position feedback device, switch-mode or linear servo amplifier, and motion controller. The integration of multiple discrete components adversely affects system reliability and renders minimization and packaging difficult particularly when multiple actuators are required. In response to this demand, a novel, low-inertia voice coil actuator has been developed whereby the traditional moving coil is replaced with a printed circuit board (PCB) that incorporates the necessary windings as conductive traces on one or more layers of the board. The result is a compact, highly integrated, highly reliable design that is simple to mass-produce using conventional PCB manufacturing and assembly techniques.

In the field of machine vision, camera calibration refers to the experimental determination of a set of parameters that describe the image formation process for a given analytical model of the machine vision system. Researchers working with low-cost digital cameras and off-the-shelf lenses generally favor camera calibration techniques that do not rely on specialized optical equipment, modifications to the hardware, or an a priori knowledge of the vision system. Most of the commonly used calibration techniques are based on the observation of a single 3-D target or multiple planar (2-D) targets with a large number of control points. This paper presents a novel calibration technique that offers improved accuracy, robustness, and efficiency over a wide range of lens distortion. This technique operates by minimizing the error between the reconstructed image points and their experimentally determined counterparts in "distortion free" space. This facilitates the incorporation of the exact lens distortion model. In addition, expressing spatial orientation in terms of unit quaternions greatly enhances the proposed calibration solution by formulating a minimally redundant system of equations that is free of singularities. Extensive performance benchmarking consisting of both computer simulation and experiments confirmed higher accuracy in calibration regardless of the amount of lens distortion present in the optics of the camera. This paper also experimentally confirmed that a comprehensive lens distortion model including higher order radial and tangential distortion terms improves calibration accuracy.