带有自锁式关节的农业六足机器人能耗优化模型及验证.

Hexapod robot has the potential to be an important agricultural equipment in hilly areas in the future, but its feature of high energy consumption has become a technical bottleneck in practical application. Energy self-sufficiency may be within our reach if the robot is driven by electric motors and can be charged by solar energy, but this type of robot often has no self-locking joints, which leads to certain energy consumption against gravity even when they stand still; although the joint brake mechanism may be helpful, but shortcomings like energy penalty and lack of compact in mechanism that come with it will appear. If the joints are self-locking to ensure no energy consumption when the robot stands still, then the energy consumption when moving will increase dramatically. Based on the above problems, to minimize overall energy consumption of the robot when it's moving or long-time standing still, a hexapod robot design with optimal disposition of self-locking joints was given, the energy consumption optimization model was also built only considering that the robot had a specific motion which avoided a non-deterministic polynomial hard problem, and thus the model just needed to optimize the energy consumption instantaneously. A defect of such an existing optimization model based on the torque distribution algorithm was corrected by introducing the difference of torque transmission efficiencies of robot joints in forward and reverse drive; in order to adapt the compacted soil ground in hilly areas, the constraints of the existing model which were suitable for the rigid environment were modified based on terrain mechanics. For different self-locking joints configurations, there would be different energy consumption optimization models, and their corresponding objective functions were given separately; this allowed energy consumption comparison of different configurations being conducted based on the minimum values of objective functions from different optimization models, and then the self-locking joints must be positioned properly and used as few as possible in the optimal configuration. In order to validate the optimization models and the effect of optimal disposition of self-locking joints, an experiment and optimization simulations were carried out. Torque transmission efficiencies of a self-locking worm joint in forward and reverse drive were tested in the experiment. The test data agreed with the ones obtained from the theoretical formula, which proved the correctness for the model to introduce the difference of torque transmission efficiency and the accurateness of the model compared with the existing one. Optimization results of the model proposed in this paper were contrasted with that of the existing one through Matlab and Adams co-simulation. Comparison results show that: joint torques as the optimum solution obtained from the optimization model of this paper have fewer chances to do negative work when robot is moving, which is better in accord with the "gravitationally decoupled actuation" and "coupled drive" concepts, and thus the total energy consumption of robot can be further reduced by more than 40% compared with the existing model. For a given robot configuration, the power consumption is more than 38 W when standing still if the robot has no self-locking joints, and the energy consumption when moving is 3 times higher than the former one if all the joints are self-locking; only when applying the optimization model and the optimal joints disposition, the robot can not only stand still with no energy consumption but also consume energy basically at the same level as the former one when moving, and the energy consumption of the robot is only 14.2%, 45.5% and 16.3% higher than that with no self-locking joints when walking downhill, uphill along a 30° ramp and level, respectively. The energy consumption optimization model proposed in the paper provides theoretical supports for the design and optimization of agricultural hexapod robot with self-locking joints. [ABSTRACT FROM AUTHOR]