18240124. ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT simplified abstract (Google LLC)
Contents
- 1 ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications of this Technology
- 1.9 Possible Prior Art
- 1.10 Unanswered Questions
- 1.11 Original Abstract Submitted
ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT
Organization Name
Inventor(s)
Catie Cuan of Stanford CA (US)
Tsang-Wei Lee of Mountain View CA (US)
Anthony G. Francis, Jr. of Simpsonville SC (US)
Alexander Toshev of San Francisco CA (US)
Soeren Pirk of Palo Alto CA (US)
ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT - A simplified explanation of the abstract
This abstract first appeared for US patent application 18240124 titled 'ROBOT NAVIGATION IN DEPENDENCE ON GESTURE(S) OF HUMAN(S) IN ENVIRONMENT WITH ROBOT
Simplified Explanation
Training and/or utilizing a high-level neural network (NN) model, such as a sequential NN model. The high-level NN model, when trained, can be used to process a sequence of consecutive state data instances (e.g., N most recent, including a current state date instance) to generate a sequence of outputs that indicate a sequence of position deltas. The sequence of position deltas can be used to generate an intermediate target position for navigation and, optionally, an intermediate target orientation that corresponds to the intermediate target position. The intermediate target position and, optionally, the intermediate target orientation, can be provided to a low-level navigation policy, such as an MPC policy, and used by the low-level navigation policy as its goal position (and optionally goal orientation) for a plurality of iterations (e.g., until a new intermediate target position (and optionally new target orientation) is generated using the high-level NN model.
- High-level neural network model trained to process consecutive state data instances
- Generates sequence of position deltas for navigation
- Intermediate target position and orientation provided to low-level navigation policy
- Used as goal position for navigation iterations
Potential Applications
- Autonomous vehicles navigation
- Robotics path planning
- Drone flight control
Problems Solved
- Efficient navigation goal setting
- Seamless integration of high-level and low-level navigation systems
Benefits
- Improved navigation accuracy
- Real-time decision-making capabilities
- Enhanced autonomous operation
Potential Commercial Applications of this Technology
Optimizing delivery routes for logistics companies
Possible Prior Art
A similar approach was used in the field of robotics for path planning algorithms in the past.
Unanswered Questions
How does the high-level NN model handle noisy or incomplete data instances during training?
The article does not address the specific methods or techniques used to handle noisy or incomplete data instances during training.
The article does not provide information on the computational resources required for training and using the high-level NN model in navigation applications.
Original Abstract Submitted
Training and/or utilizing a high-level neural network (NN) model, such as a sequential NN model. The high-level NN model, when trained, can be used to process a sequence of consecutive state data instances (e.g., N most recent, including a current state date instance) to generate a sequence of outputs that indicate a sequence of position deltas. The sequence of position deltas can be used to generate an intermediate target position for navigation and, optionally, an intermediate target orientation that corresponds to the intermediate target position. The intermediate target position and, optionally, the intermediate target orientation, can be provided to a low-level navigation policy, such as an MPC policy, and used by the low-level navigation policy as its goal position (and optionally goal orientation) for a plurality of iterations (e.g., until a new intermediate target position (and optionally new target orientation) is generated using the high-level NN model.