Reinforcement Learning is a commonly used technique for learning tasks in robotics, however, traditional algorithms are unable to handle large amounts of data coming from the robot’s sensors, require long training ti...Reinforcement Learning is a commonly used technique for learning tasks in robotics, however, traditional algorithms are unable to handle large amounts of data coming from the robot’s sensors, require long training times, and use dis-crete actions. This work introduces TS-RRLCA, a two stage method to tackle these problems. In the first stage, low-level data coming from the robot’s sensors is transformed into a more natural, relational representation based on rooms, walls, corners, doors and obstacles, significantly reducing the state space. We use this representation along with Behavioural Cloning, i.e., traces provided by the user;to learn, in few iterations, a relational control policy with discrete actions which can be re-used in different environments. In the second stage, we use Locally Weighted Regression to transform the initial policy into a continuous actions policy. We tested our approach in simulation and with a real service robot on different environments for different navigation and following tasks. Results show how the policies can be used on different domains and perform smoother, faster and shorter paths than the original discrete actions policies.展开更多
The overall research in Reinforcement Learning (RL) concentrates on discrete sets of actions, but for certain real-world problems it is important to have methods which are able to find good strategies using actions dr...The overall research in Reinforcement Learning (RL) concentrates on discrete sets of actions, but for certain real-world problems it is important to have methods which are able to find good strategies using actions drawn from continuous sets. This paper describes a simple control task called direction finder and its known optimal solution for both discrete and continuous actions. It allows for comparison of RL solution methods based on their value functions. In order to solve the control task for continuous actions, a simple idea for generalising them by means of feature vectors is presented. The resulting algorithm is applied using different choices of feature calculations. For comparing their performance a simple measure is展开更多
The right to claim for damages for infringement is of the character of credit and an object of limitation of action. In case of trademark right infringement, the provision on limitation of action of the General Pr... The right to claim for damages for infringement is of the character of credit and an object of limitation of action. In case of trademark right infringement, the provision on limitation of action of the General Principles of the Civil Law also apply to the trademark proprietor's right to claim for damages for infringement.……展开更多
We introduce notions of continuous orbit equivalence and its one-sided version for countable left Ore semigroup actions on compact spaces by surjective local homeomorphisms,and characterize them in terms of the corres...We introduce notions of continuous orbit equivalence and its one-sided version for countable left Ore semigroup actions on compact spaces by surjective local homeomorphisms,and characterize them in terms of the corresponding transformation groupoids and their operator algebras.In particular,we show that two essentially free semigroup actions on totally disconnected compact spaces are continuously orbit equivalent if and only if there is a canonical abelian subalgebra preserving C^(∗)-isomorphism between the associated transformation groupoid C^(∗)-algebras.We also give some examples of orbit equivalence,consider the special case of semigroup actions by homeomorphisms and relate continuous orbit equivalence of semigroup actions to that of the associated group actions.展开更多
Most studies on solute transport in coastal aquifers affected by tides focus on the transport of instantaneous released solute,and there are few studies on continuously released solute affected by tides.In this study,...Most studies on solute transport in coastal aquifers affected by tides focus on the transport of instantaneous released solute,and there are few studies on continuously released solute affected by tides.In this study,the image monitoring method is used to establish the quantitative relationship between the concentration of the colored tracer and the hue value of the image,and the digital image is used to determine the tracer concentration distribution.Using image monitoring method laboratory experiments,quantitative analysis of the characteristics of continuously released solute transport in coastal unconfined aquifers under the tidal influence.Experiments show that the high tide inhibits the increase in the concentration of each point in the aquifer.Under the influence of tides,the solute plume retreats towards the land.During the low tide period,the solute plume migrates toward the sea again.And the solute plume will maintain a relatively stable shape after entering the aquifer for a long enough time.Ignoring the tidal effect seems to have little effect on the estimation of the position of the solute plume,but ignoring the tidal effect has a certain influence on the estimation of the dispersion range of the solute plume.No matter whether considering the tidal action,the final dispersion range of the solute plume is almost the same.But before the solute plume reaches a stable state,ignoring the tidal effect will lead to a smaller dispersion range of the solute plume.展开更多
文摘Reinforcement Learning is a commonly used technique for learning tasks in robotics, however, traditional algorithms are unable to handle large amounts of data coming from the robot’s sensors, require long training times, and use dis-crete actions. This work introduces TS-RRLCA, a two stage method to tackle these problems. In the first stage, low-level data coming from the robot’s sensors is transformed into a more natural, relational representation based on rooms, walls, corners, doors and obstacles, significantly reducing the state space. We use this representation along with Behavioural Cloning, i.e., traces provided by the user;to learn, in few iterations, a relational control policy with discrete actions which can be re-used in different environments. In the second stage, we use Locally Weighted Regression to transform the initial policy into a continuous actions policy. We tested our approach in simulation and with a real service robot on different environments for different navigation and following tasks. Results show how the policies can be used on different domains and perform smoother, faster and shorter paths than the original discrete actions policies.
文摘The overall research in Reinforcement Learning (RL) concentrates on discrete sets of actions, but for certain real-world problems it is important to have methods which are able to find good strategies using actions drawn from continuous sets. This paper describes a simple control task called direction finder and its known optimal solution for both discrete and continuous actions. It allows for comparison of RL solution methods based on their value functions. In order to solve the control task for continuous actions, a simple idea for generalising them by means of feature vectors is presented. The resulting algorithm is applied using different choices of feature calculations. For comparing their performance a simple measure is
文摘 The right to claim for damages for infringement is of the character of credit and an object of limitation of action. In case of trademark right infringement, the provision on limitation of action of the General Principles of the Civil Law also apply to the trademark proprietor's right to claim for damages for infringement.……
基金Supported by the NSF of China(Grant No.12271469,11771379,11971419)。
文摘We introduce notions of continuous orbit equivalence and its one-sided version for countable left Ore semigroup actions on compact spaces by surjective local homeomorphisms,and characterize them in terms of the corresponding transformation groupoids and their operator algebras.In particular,we show that two essentially free semigroup actions on totally disconnected compact spaces are continuously orbit equivalent if and only if there is a canonical abelian subalgebra preserving C^(∗)-isomorphism between the associated transformation groupoid C^(∗)-algebras.We also give some examples of orbit equivalence,consider the special case of semigroup actions by homeomorphisms and relate continuous orbit equivalence of semigroup actions to that of the associated group actions.
基金supported by the National Natural Science Foundation of China(No.42172281)the Opening Fund of the State Key Laboratory of China University of Geosciences(Wuhan)(No.SKJ2018055)。
文摘Most studies on solute transport in coastal aquifers affected by tides focus on the transport of instantaneous released solute,and there are few studies on continuously released solute affected by tides.In this study,the image monitoring method is used to establish the quantitative relationship between the concentration of the colored tracer and the hue value of the image,and the digital image is used to determine the tracer concentration distribution.Using image monitoring method laboratory experiments,quantitative analysis of the characteristics of continuously released solute transport in coastal unconfined aquifers under the tidal influence.Experiments show that the high tide inhibits the increase in the concentration of each point in the aquifer.Under the influence of tides,the solute plume retreats towards the land.During the low tide period,the solute plume migrates toward the sea again.And the solute plume will maintain a relatively stable shape after entering the aquifer for a long enough time.Ignoring the tidal effect seems to have little effect on the estimation of the position of the solute plume,but ignoring the tidal effect has a certain influence on the estimation of the dispersion range of the solute plume.No matter whether considering the tidal action,the final dispersion range of the solute plume is almost the same.But before the solute plume reaches a stable state,ignoring the tidal effect will lead to a smaller dispersion range of the solute plume.
