This paper presents a general procedure for enabling autonomous row following in crops during early-stage \ngrowth, without relying on absolute localization systems. A model based on deep learning techniques (object \ndetection for wide-row crops and segmentation for narrow-row crops) was applied to accurately detect both \ntypes of crops. Tests were performed using a manually operated mobile platform equipped with an RGB and \na time-of-flight (ToF) cameras. Data were acquired during different time periods and weather conditions, in \nmaize and wheat fields. The results showed the success on crop detection and enables the future development \nof a fully autonomous navigation system in cultivated fields during early stage of crop growth.

Research on legged locomotion has encountered its limits in human-made actuation technology. Building a structure and an actuation system capable of performing like a biological muscle for the dynamic locomotion of a mid- to large-sized machine is prohibitive. Novel technologies are being explored with the aim of improving actuator performance. This paper describes the development of a robotic leg for agile locomotion. The main aim is to imitate the multidisciplinary performance of the natural muscle by means of hybridizing conventional and new technologies for the development of large power, large force, lightweight, force-controlled, compliant robotic legs. A preliminary leg prototype has been developed. The actuation system is based on the hybrid use of series elasticity and magneto-rheological dampers which provides variable compliance for natural-looking motion and improved interaction with the ground. Experiments with the real leg prototype provide a positive assessment of the target leg features.

In this paper we present a new method for measuring forces using piezoresistive sensors. This method dramatically increases the accuracy and repeatability of the readings compared with traditional methods. These improvements will allow the common use of piezoresistive sensors in robotic applications, where only expensive and sophisticated force sensors such as load cells are currently used. We also present a sketch of a haptic interface under development, consisting of a data glove in which these force sensors are integrated. This interface is intended to control an Intelligent Assist Device by measuring operator-applied forces to the load. The new method proposed for measuring forces in piezoresistive sensors consists of reading conductance and capacitance by applying DC and sinusoidal waveforms to the sensors, thereby allow us to determine a multivariable estimation of force, instead of using the traditional, purely resistive model that has been used up to now.

It is a widespread idea that animal-legged locomotion is better than wheeled locomotion on natural rough terrain. However, the use of legs as a locomotion system for vehicles and robots still has a long way to go before it can compete with wheels and trucks, even on natural ground. This paper aims to solve two main disadvantages plaguing walking robots: their inability to react to external disturbances (which is also a drawback of wheeled robots); and their extreme slowness. Both problems are reduced here by combining: 1) a gait-parameter-adaptation method that maximizes a dynamic energy stability margin and 2) an active-compliance controller with a new term that compensates for stability variations, thus helping the robot react stably in the face of disturbances. As a result, the combined gait-adaptation approach helps the robot achieve faster, more stable compliant motions than conventional controllers. Experiments performed with the SILO4 quadruped robot show a relevant improvement in the walking gait.

Walking-robot technology has reached an advanced stage of development, as has already been demonstrated by a number of real applications. However, further improvement is still needed if walking robots are to compete with traditional vehicles. Some potential improvements could be gained through optimization. Thus, this paper presents a method for distributing the legs around the robot’s body such as to reduce the forces the legs must exert to support and propel the robot. The method finds through non-linear optimization techniques the middle leg displacement that nulls the difference between foot forces in a middle leg and a corner leg. A walking robot has been built to assess the theoretical results.

This paper presents a method for generating free gaits for quadruped robots capable of performing statically stable, omnidirectional locomotion on irregular terrain containing forbidden areas. The rule-based deliberative algorithm can generate flexible sequences of leg transferences while maintaining constant vehicle speed. The foothold planning method is compatible with the use of these flexible leg sequences, and is designed to maintain a minimum absolute stability margin despite the terrain height uncertainty. The integration of exteroceptive terrain profile data has been considered to improve adaptability. Experimental results are presented to show the gait's efficiency in adapting to an irregular terrain containing forbidden areas.

Legged locomotion has been extensively investigated in the last two decades, during which time many walking machines have been built as test-beds for research purposes. Legged locomotion technology now seems to be ready to take on industrial applications. This paper presents a legged machine-ROWER-for a real industrial application in naval construction processes. The main task of the walking machine is to provide full motion to a welding manipulator in a very complex scenario, the double hull of a ship. This scenario consists of cells fitted with stiffeners on the top, bottom, and lateral walls to reinforce the ship's structure. The legged machine moves through the cell grasping the top and bottom reinforcements rather than walking on the bottom plane. This improves the machine's stiffness and guarantees its stability. This paper describes the walking machine's main structure, the leg structure, and grasping mechanism. It then defines and simulates the gait for walking while grasping the stiffeners. Finally, it reports the features of a full-size leg.