Early robotic grasping tasks primarily rely on cooperative target information, where robots follow pre-programmed control routines, mechanically repeating a series of fixed actions. This grasping operation not only has low efficiency but also has limited adaption for specific scenarios.
Robotic grasping generation is a prerequisite for grasping operations. Depending on the application scenario, it involves target recognition, target pose estimation and grasping force estimation. In this stage, the robot perceives target feature information to determine the grasping pose and grasping force of the target.
In recent years, there have been numerous vision-based methods proposed for target recognition in robotic grasping. However, when dealing with visually indistinguishable objects or occluded scenes, which can lead to challenges in target recognition.
In robotics, solar cells are increasingly being used as a renewable, stable and autonomous power source for smaller robots, and as photovoltaic (PV) technology continues to progress, they can be expected to power larger and even humanoid robots in the future. The most common power sources for robots at present are integrated batteries (Kaur, 2019).
Solar cells are integrated into the robot’s chassis and power is generated for the robot through the photovoltaic effect. When selecting solar cells for onboard power generation, factors such as the efficiency of the surface area, efficiency to weight, economic cost and durability must be considered.
To ensure a stable grasp, some researchers have proposed adjustment strategies based on tactile information to address unstable states. Their aim is to use real-time tactile or tactile-visual fusion feedback information to adjust the robot's grasp pose and force, adapting to the instability or changes in the target object.