Search results

Automatic target recognition in agricultural harvesting robots is characterized by low detection rates and high false alarm rates due to the unstructured nature of both the environment and the objects. To improve detection human‐robot collaboration levels were defined and implemented. The collaboration level is defined as the level of system autonomy or the level at which the human operator (HO) interacts with the system. Experimental results on images taken in the field indicate that collaboration of HO and robot increases detection and reduces the time required for detection.

To follow and maintain an appropriate distance to the selected target person (STP), the mobile robot is required to have capabilities: the human detection and tracking and an efficient following strategy with a smooth manner that does not appear threatening to the STP and surroundings. The efficient following strategy must integrate the STP position and the obstacle information to achieve smooth and safe human-following behaviors, especially in unknown environments where robot does not have understandings in advance. The purpose of this study is to propose a robust-adaptive-behavior strategy for mobile robots.

This paper presents a robust-adaptive-behavior strategy (RABS) based on the fuzzy inference mechanism to help the robot follow the STP effectively in various unknown environments with the real-time obstacle avoidance, both indoor and outdoor and on different robot platforms. In which, the traversability of robots’ unknown surrounding environments is analyzed by using the STP position and the obstacle information obtained from the two dimensional laser scan, whose purpose is to choose the highest-traversability-score direction (HTSD) and an adaptive-safe-following distance (ASFD). Then, the HTSD, the ASFD and the current velocity of the robot are considered as inputs of the fuzzy system to adjust its velocity smoothly.

The proposed RABS is verified by a set of experiments using a real big-heavy autonomous mobile robot (BH-AMR), with the dimension 0.8 × 1.2 (m), weight 150 (kg), full-load 500 (kg), aiding smart factories. The obtained results have shown that the proposed RABS equips the BH-AMR with the ability to follow the STP smoothly and safely even when the robot is moving at the maximum speed 1.5 (m/s).

In this paper, the autonomous mobile robot considers all environments as unknown even when it is working in mapped environments. This limitation is presented clearly in the future works section.

This proposed method can be used to help the autonomous mobile robot support persons in factories, hospitals, restaurants, supermarkets or at the airports.

This paper presents a RABS, including three new features: a fuzzy-based solution to help human-following robots maintain an appropriate distance to the STP safely and smoothly with the maximum velocity 1.5 (m/s); the proposed fuzzy-based solution, an adaptive vector field histogram and a new approach for the STP tracking is combined to follow the STP and avoid the collision simultaneously in unknown indoor and outdoor environments; the proposed RABS is considered for BH-AMRs (with the dimension 0.8 × 1.2 (m), weight 150 (kg), full-load 500 (kg)) to serve real tasks in smart factories.